archive-voice

[vocal valley]

This channel is for discussing our new voice library: https://github.com/discordjs/voice — we're excited to hear your thoughts and any questions/issues you have! 🎉

📌 What is @discordjs/voice?
This is a new voice library that we're working on that aims to be more reliable, scalable, and offer better playback quality than what we currently offer in Discord.js v12.

📌 What is happening to my existing voice code that uses Discord.js v12?
From v13, we are dropping support for the existing voice codebase to focus on the new voice library instead. The new voice library may look more intimidating, but we have created a thought-out music bot example (see examples below) to help you transition your code. Using our example music player means you'll likely end up writing less voice code in your bot!

📌 How do I use this with libraries that aren't discord.js?
You can use the new voice library anywhere you want by creating your own custom adapter. If you want to use the library with discord.js, there is currently a PR in progress (https://github.com/discordjs/discord.js/pull/5402) to add the required adapters directly to discord.js so you won't have to create them yourself. In the meantime, the examples in the voice repository work with v12.

📌 What about voice receive?
This feature isn't documented in the Voice API by Discord and is prone to breaking quite a lot. We have basic support for it in the voice library currently, but there is still work going on to add utilities to make it easier to work with.

📌 Useful links:

• GitHub Repo: https://github.com/discordjs/voice
• Examples: https://github.com/discordjs/voice/tree/main/examples
• GitHub Discussions: https://github.com/discordjs/voice/discussions (e.g. to give feedback on the library or to hold longer discussions that can be revisited in the future)
• Guide: https://discordjs.guide/voice/

Yes, this is a new channel! But please don't spam otherwise you'll be moderated and restricted 👍

[plain vine]

are common bugs like having to pause resume pause resume still there? also ty hydra o/

[fervent estuary]

all of those are gone

[vocal valley]

they should not be anymore

[plain vine]

great!

[next herald]

Will this affect me if I'm using erela.js for music? But DJS for the rest of my bot ?

[steel crescent]

It will affect you as you can’t really detect a users voice state easily anymore

But if you use erela, it wouldn’t matter to much I’d think

Since almost everything is handled by Lavalink anyway by you forwarding the events

(Which we obv still allow)

[next herald]

Thanks for info!

[cold geode]

What are the advantages of this innovation?

[vocal valley]

• playback quality is improved
• you can use it with libraries other than discord.js
• more reliable (in general)
• you can scale it horizontally, run your voice in a separate process from your bot

[cold geode]

thank you

[faint forge]

oh nice

[vocal valley]

Just updated the pinned message, but if anyone wanted a high level overview of the new voice library, there is a slightly outdated guide available here https://deploy-preview-595--discordjs-guide.netlify.app/voice/. With the exception of voice adapters, everything else should still hold (oh and we are not removing voice receive like that guide says) 🙂

[haughty cape]

The master branch hasnt yet dropped the in-built support, right?

[carmine timber]

Ok

I got it

[vocal valley]

👍

Since you aren't using typescript, you might find it easier to copy the adapter file from here: https://github.com/amishshah/discord-radio-bot/blob/feat/upgrade/src/adapter/adapter.js

Once the PR is merged, this file will no longer be needed

[carmine timber]

I thought adaptor Creator is not required since join voice channel .ts has nothing to mention in it.

[pallid flint]

Why did the devs thought to remove guild.member lol

[radiant ocean]

that's not the question for this channel, and the PR that did that explained why

it is literally nothing more than guild.members.resolve()

[vocal valley]

can you show me where this is? because it definitely is required

[warm pebble]

question will discordjs/voice be supported for discord.js v12 or is it for v13 only?

[carmine timber]

Sorry My bad

I got it

LOL

[vocal valley]

it will work with v12 also 🙂 we have two examples on the repo - the basic one works with v12, and there is a music bot example that uses the @zealous hearth branch of discord.js (v13) but that is only for slash commands, you can easily make it work with v12

[warm pebble]

alright thanks!

[vocal valley]

the idea of this library is that it will work anywhere where you have a discord gateway connection, so it could even work with other libraries such as eris

[warm pebble]

cool

so in the future #archive-djs-v12-voice-deprecated will no longer be supported i guess

[vocal valley]

yep that is the plan, this channel will eventually become the only voice support channel

[warm pebble]

alright, thanks for the info

[vocal valley]

np 😄

[zealous hearth]

But why did you pinged me......

[vocal valley]

oops sorry about that 👀

[zealous hearth]

Understandable

[barren flume]

is this compatible with erela?

[vocal valley]

afaik erela is a way to use lavalink with your discord.js bot? in that case, no, since lavalink handles voice stuff, not @discordjs/voice

if you are using erela this library should not affect you

[barren flume]

so this is specefically designed for stuff much like ytdl ?

[vocal valley]

you can definitely use ytdl with this library 🙂 it's more targeted towards people who currently use discord.js v12 voice

[barren flume]

k thanks 😄

[carmine timber]

@vocal valley
Now is this ok ??

const { joinVoiceChannel } = require('@discordjs/voice')
const adaptor = require('../util/adaptor')

exports.run = (client, message, args) => {
    const channel = message.channel
    var connection =  joinVoiceChannel({
        channelId: channel.id,
        guildId: channel.guild.id,
        adapterCreator : adaptor.createDiscordJSAdapter(message.member.voice.channel)
    })
    console.log(connection)
}

I get a signalling status but bot never joins vc

[vocal valley]

hmm, does your bot have permission to do so?

[carmine timber]

It has administrator access

What else he needs ??

[tulip jackal]

you're trying to connect to a text channel

[carmine timber]

Oh

May be

XD

[carmine timber]

Thanks it worked

[split cliff]

Does the voice in v13 mean we can use voice-recognition?

[vocal valley]

if you mean recognising what people say in a voice channel, then maybe - it depends on how stable voice receive is. But there is nothing stopping you from doing something similar with the voice in v12 anyway

[split cliff]

Hmn, okay ty

[vocal valley]

this might be useful for you to have a look at, https://github.com/devsnek/scribe

[wind ginkgo]

Would this affect any JS Lavalink libraries? I use Lavaclient, so I'm not sure if this will affect anything or not

[vocal valley]

nope, only affects those who use discord.js's built-in voice currently

[swift forum]

Native, as in.. C or Rust?

[carmine timber]

Yes I am successful to connect to vc and play a stream with new audio api

[vocal valley]

amazing! 🎉 if you have feedback would love to receive it, either here on github discussions (would prefer on github discussions!)

[vocal valley]

sorry that was a misleading choice of words, i think "first class support" is more apt here

discord.js just exposes some of its internals so that you can use it with @discordjs/voice without as much boilerplate code

[swift forum]

Ah, I see!

[carmine timber]

It sounds very clear and stable at same time. It is a huge upgrade over existing discord v12 voice api. Performance has hugely been increased and stability of voice connection of stream is also very good. Discord js was already very fast, but after this voice api, I can see a huge difference in performance when playing song for first time.

Huge Thanks to developers for this new Voice API

[vocal valley]

this is very nice to hear, thanks for the appreciation 😊

[drowsy lynx]

it is stage channel thing?

[vocal valley]

this library does support stage channels, yes

[drowsy lynx]

oh

[subtle granite]

whats this about

[fervent estuary]

check the pinned message

[carmine timber]

@vocal valley
Can you explain ?? Your docs told me to use connection for pause and unpause

But I needed to digg and get player from connection and then use pause and unpause command.

Docs for pausing and unpausing : https://deploy-preview-595--discordjs-guide.netlify.app/voice/audio-player.html#pausing-unpausing

[vocal valley]

@carmine timber sorry about that, that's one of the outdated parts of that particular documentation

[carmine timber]

Ok no problem 🙂

[vocal valley]

you should use audioPlayer.pause() and unpause instead, as you have said

[carmine timber]

But I am using command handler and audio player is created in play command while pause and unpause is used in different command handler

So I have to search for player in connection

[vocal valley]

so when you call connection.subscribe(player), you get a subscription object which has .player and .connection properties,

you can either save this subscription object somewhere, or do something like this:

if (connection.state.status !== VoiceConnectionStatus.Destroyed) {
  connection.state.subscription?.player.pause();
}

if you are using javascript, you can trim it down a little bit by removing the if() {} around that line of code

[carmine timber]

I did this only 🙂

[vocal valley]

yep, i think that is a good way to do it

[subtle granite]

are there docs for the new voice lib?

[vocal valley]

yep! https://discordjs.github.io/voice/ I recommend reading the documentation alongside an example (https://github.com/discordjs/voice/tree/main/examples)

[subtle granite]

thank you so much!

[vocal valley]

np 😄

[vocal cove]

what makes djs-voice better than lavalink ?

[vocal valley]

I would say that @discordjs/voice is easier to integrate into a bot than lavalink and can also run in the same process as your bot, which is good depending on your use case

It also offers the developer more chance to interact with the Voice API at a lower level

[vocal cove]

My point is more about how would it be better in term of reliability/scalability/performance and load balance

[wind ginkgo]

I wonder what the performance difference will be if you benchmark the two against each other

[fresh wagon]

nice

i want to try this out soon

[carmine timber]

I have tested This XD

[austere sphinx]

are there examples in regular js instead of typescript?

[carmine timber]

It will come soon in docs

[austere sphinx]

ok

[stark drum]

well to get the js version just remove anything that's specific to typescript

[carmine timber]

You can use tsc to convert it to js

[steel crescent]

Tbf not everyone has the use case to have huge bots playing music at such a capacity. If you want something fast and non-complicated (compared to hosting your own lavalink, setting it up etc) this is def the way to go

We can’t and certainly won’t convince you to use one over the other. If one solution works for you, stick with it 👍

[subtle granite]

Whats new for voice?

[hazy copper]

just started converting my old one and i cant seem to be able to create a voice connection? im getting this error

[shut path]

is the new voice library added to discord.js@zealous hearth?

reeeeeeeeee

[subtle granite]

V13*

[carmine timber]

You need a adapter module for that

https://github.com/amishshah/discord-radio-bot/tree/feat/upgrade/src/adapter

Use js version

Then do something like this :

const { joinVoiceChannel } = require('@discordjs/voice')
const adaptor = require('<your adapter file path>')
   
//Use this below command in message event
var connection =  joinVoiceChannel({
        channelId: message.member.voice.channel.id,
        guildId: message.guild.id,
        adapterCreator : adaptor.createDiscordJSAdapter(message.member.voice.channel),
    })

[hazy copper]

looks like that "fixes the error" but the bot still doesn't join my channel

[carmine timber]

Show me your code

[hazy copper]

its as basic as it gets rn

const { joinVoiceChannel } = require('@discordjs/voice');
const adaptor = require('../../bin/adapter');

module.exports = {
    name: 'test',
    description: 'test',
    execute(interaction) {
        const connection = joinVoiceChannel({
            channelId: interaction.member.voice.channel,
            guildId: interaction.guild.id,
            adapterCreator: adaptor.createDiscordJSAdapter(interaction.member.voice.channel)
        });
    },
};

not using my main command yet so i can make this work

[carmine timber]

are you using interaction ??

[fervent jacinth]

but adapter is in discord.js in docs https://discord.js.org/#/docs/main/master/class/Guild?scrollTo=voiceAdapterCreator

[hazy copper]

yes

[carmine timber]

Try to use it with normal command

and then switch to slash

[hazy copper]

i mean why wouldn't it work with an interaction? it gets the same voice channel object it would get anyway?

[carmine timber]

I know that but I have some doubts over interaction

[hazy copper]

i dont think that's the issue, will try it in my message event

[carmine timber]

Do it and tell me

[hazy copper]

idk why i even tried this, it does the same exact thing

[carmine timber]

You are not getting id in channel id

idiot

See my code

[hazy copper]

ah it needs it's id? lulw

it should print an error for that

[carmine timber]

LOL

[carmine timber]

did it connect ??

[hazy copper]

yeah

why hasn't the adapter been merged yet tho

[carmine timber]

IDK

[vocal valley]

Merged but not released yet

[carmine timber]

Oh

[hazy copper]

will i have it if i update? (im on the github branch)

[carmine timber]

Which class will it belong to ??

[vocal valley]

Should be a new version of the master branch released in the next few minutes

[vocal valley]

Guild

[carmine timber]

Has docs been updated for that ??

[vocal valley]

The docs for master branch should be updated to show that

Adapters are now published in https://www.npmjs.com/package/discord.js/v/13.0.0-dev.c6aeebb18d6b969f7c8bdb1b719883d4384dd03e

[carmine timber]

Thanks

[vocal valley]

https://discord.js.org/#/docs/main/master/class/Guild?scrollTo=voiceAdapterCreator

[carmine timber]

2021-06-09T17:27:40.526365+00:00 app[worker.1]: /app/node_modules/@discordjs/voice/dist/networking/Networking.js:413
2021-06-09T17:27:40.526401+00:00 app[worker.1]:                 secretbox.methods.close(opusPacket, connectionData.nonceBuffer, secretKey),
2021-06-09T17:27:40.526402+00:00 app[worker.1]:                                   ^
2021-06-09T17:27:40.526402+00:00 app[worker.1]: 
2021-06-09T17:27:40.526403+00:00 app[worker.1]: TypeError: secretbox.methods.close is not a function
2021-06-09T17:27:40.526404+00:00 app[worker.1]:     at Networking.encryptOpusPacket (/app/node_modules/@discordjs/voice/dist/networking/Networking.js:413:35)
2021-06-09T17:27:40.526404+00:00 app[worker.1]:     at Networking.createAudioPacket (/app/node_modules/@discordjs/voice/dist/networking/Networking.js:397:53)
2021-06-09T17:27:40.526404+00:00 app[worker.1]:     at Networking.prepareAudioPacket (/app/node_modules/@discordjs/voice/dist/networking/Networking.js:321:37)
2021-06-09T17:27:40.526405+00:00 app[worker.1]:     at VoiceConnection.prepareAudioPacket (/app/node_modules/@discordjs/voice/dist/VoiceConnection.js:282:33)
2021-06-09T17:27:40.526405+00:00 app[worker.1]:     at /app/node_modules/@discordjs/voice/dist/audio/AudioPlayer.js:419:54
2021-06-09T17:27:40.526405+00:00 app[worker.1]:     at Array.forEach (<anonymous>)
2021-06-09T17:27:40.526406+00:00 app[worker.1]:     at AudioPlayer._preparePacket (/app/node_modules/@discordjs/voice/dist/audio/AudioPlayer.js:419:19)
2021-06-09T17:27:40.526406+00:00 app[worker.1]:     at AudioPlayer._stepPrepare (/app/node_modules/@discordjs/voice/dist/audio/AudioPlayer.js:391:22)
2021-06-09T17:27:40.526406+00:00 app[worker.1]:     at prepareNextAudioFrame (/app/node_modules/@discordjs/voice/dist/DataStore.js:71:31)
2021-06-09T17:27:40.526407+00:00 app[worker.1]:     at audioCycleStep (/app/node_modules/@discordjs/voice/dist/DataStore.js:56:5)

Crashing after update, I think Audio Player is messed up again

[vocal valley]

could you do this for me:

[carmine timber]

What ??

[vocal valley]

import generateDependencyReport from the voice library, and log the value of it after you call it

console.log(generateDependencyReport ())

[carmine timber]

Ok wait

[vocal valley]

That error indicates you might be missing sodium, tweetnacl, or libsodium-wrappers (you need at least one)

[hazy copper]

is selfdeaf supposed to be on by default?

[carmine timber]

I am getting this

Sorry

That was wrong

[vocal valley]

Np

[carmine timber]

--------------------------------------------------
Core Dependencies
- @discordjs/voice: 0.3.0
- prism-media: 1.2.9

Opus Libraries
- @discordjs/opus: 0.5.0
- opusscript: not found

Encryption Libraries
- sodium: not found
- libsodium-wrappers: not found
- tweetnacl: 1.0.3

FFmpeg
- version: 2021-01-12-git-ca21cb1e36-full_build-www.gyan.dev
- libopus: yes
--------------------------------------------------

[vocal valley]

hmm interesting, I will take a look

could you try installing libsodium-wrappers instead

[carmine timber]

Ok

[carmine timber]

Can I use tweetnacl

??

[vocal valley]

You should be able to, but I want to see if the error is because of tweetnacl

[carmine timber]

Ok

[vocal valley]

So if you install libsodium-wrappers and it works I can see what I need to change for tweetnacl 👍

[carmine timber]

It works with libsodium

[vocal valley]

ok that's good, I will try to see why it doesn't work with tweetnacl, thanks for finding this

[carmine timber]

Which is faster ? libsodium or tweetnacl

[vocal valley]

Libsodium-wrappers

And even faster is sodium

[carmine timber]

Ok thanks

[torn sinew]

Hi,
so with old ones I was trying to write up a thing that would do the following

1. Play Background music
2. Receive a command
3. fade out the background (slowly reduce the volume)
4. Play a sound
5. Resume the background with a fade in (slowly increasing the volume)

But it proved troublesome... from a brief look at the docs I see that as a starter the volume is controlled on the resource level not player level, so probably no dice. But man can dream :D.

[carmine timber]

selfdeaf is on by default

[carmine timber]

You can use sleep command for that and after that loop command each second or each 0.5 second and every time you loop, you are reducing volume of dispatcher. And the same goes for next song.

Example of sleep : https://stackoverflow.com/questions/951021/what-is-the-javascript-version-of-sleep

@vocal valley For some reasons, after installing libsodium-wrappers, it still shows me this :

--------------------------------------------------
Core Dependencies
- @discordjs/voice: 0.3.0
- prism-media: 1.2.9

Opus Libraries
- @discordjs/opus: 0.5.0
- opusscript: not found

Encryption Libraries
- sodium: not found
- libsodium-wrappers: not found
- tweetnacl: 1.0.3

FFmpeg
- version: 2021-01-12-git-ca21cb1e36-full_build-www.gyan.dev
- libopus: yes
--------------------------------------------------

AS Not Found.

[vocal valley]

hmm

is that even after you restart your bot script?

[carmine timber]

Yes

[vocal valley]

how odd

can you show me your package.json?

[carmine timber]

Ok wait

{
  "name": "vc-check",
  "version": "1.0.0",
  "main": "index.js",
  "scripts": {
    "test": "echo \"Error: no test specified\" && exit 1"
  },
  "author": "",
  "license": "ISC",
  "dependencies": {
    "@discordjs/opus": "^0.5.0",
    "@discordjs/voice": "^0.3.0",
    "discord.js": "^13.0.0-dev.c6aeebb18d6b969f7c8bdb1b719883d4384dd03e",
    "install": "^0.13.0",
    "libsodium-wrappers": "^0.7.9",
    "npm": "^7.16.0",
    "youtube-sr": "^4.1.4",
    "ytdl-core-discord": "^1.3.0"
  },
  "devDependencies": {},
  "description": ""
}

See this

@vocal valley ??

[vocal valley]

yep

i'll take a look

[carmine timber]

Ok

[vocal valley]

will look into it 👍

[carmine timber]

👍

[crude parcel]

:new-voice: 🤔

[opal mauve]

when trying to compile https://i.imgur.com/sZ7aHbu.png

[final harbor]

Enable esModuleInterop in the config should do the trick

But it's that flag causing it tho

[frosty quartz]

Heya , i see that wonderfull new channel for voice and the new api that Hydra made as well and i was wondering : I already have a complete music bot , functioning pretty well and i would like to change from YTDL to djs voice one day or another. Is that possible to do every command that i already have with that api ? ( ping me when answer please i'll be at work so i'll check the message in about 7 hours.) Thanks in advance 😄 ❤️

[carmine timber]

Codes are different for playing but YTDL will still work for you in new Voice API

[rancid heart]

When should I switch to the new library?

[carmine timber]

Let the user-friendly documentation come out and then switch easily

[upper flare]

i would assume that ytdl is needed regardless.

[spring jackal]

Have there been any updates to voice? I'm using code that worked in v12 suddenly after the update I keep getting 'VOICE_CONNECTION_TIME" ...

[vocal valley]

Hi Lilly, you should be able to do all the same things you could do with discord.js V12 voice with the new voice library too 😁 our example code has a music bot with a queue, play, resume and stop too

[vocal valley]

I would recommend at least having a look at the music bot example now to see what the upgrade will look like and then seeing if you feel comfortable using the examples and documentation to upgrade now. If you are feeling intimidated by the documentation, I'm also writing additional documentation that is a bit more user friendly

[vocal valley]

I think you're looking for #archive-djs-v12-voice-deprecated, but nothing has changed to the voice on v12 - might be an issue with your network

[spring jackal]

no I mean I updated to v13 and now I'm getting that error

[vocal valley]

if you update to the newest version of the Dev release, voice is removed and we use this new library instead (see pinned messages)

[spring jackal]

Is this how it will work on v13 or is this some kind of test project to implement to v13?

[vocal valley]

This is how it will work on v13, yep

[vocal cove]

that's like reinventing the wheel, we could just recommend lavalink and spare the long/hard work on djs voice, and just focus on the main djs to make it better.
in the other hand having a djs voice that follow the same rules and conventions like the rest of djs, would be definitely easier for a starter when starting out with djs
my point is, if you're not making it better, don't waste much time and effort on it on remaking it, and waste your users time and effort on upgrading to it
this is just an opinion tho, no offense intended
thanks

[sacred cove]

this library hasn't got anything to do with @discordjs/opus and it's still required isn't it?

[steel crescent]

You say that as if lavalink has been around for ages and we should tell everyone to have a proper host to deploy a jar or potentially docker instance with lavalink, which in its idle state eats around 200MB+ RAM already.
Lavalink is a service, /voice is a tool. You or anyone can replicate lavalink easily with /voice.

[vocal valley]

well it is related to @discordjs/opus, as the opus library is a dependency of this one

[sacred cove]

oh I see, and is it required?

no wait that's a stupid question

[vocal valley]

nope, you could use opusscript instead

[sacred cove]

oh ok

[vocal cove]

so you're saying that it won't each as much ram as Lavalink, this is already a good start.
this is exactly what I meant, I did ask of what would make discordjs voice, there was no answer on that earlier

having various points where discordjs voice score better than lavalink or other libraries would a good way on presenting it

[vocal valley]

it's not an easy question to give an answer to because atm there are no comparisons that we've made

[vocal cove]

yea it's still under the work I know

[vocal valley]

I would definitely recommend @discordjs/voice if you're doing something novel, e.g. making a game with like voiceovers

[vocal cove]

but it has to be a goal on remaking what is already made

[vocal valley]

for the typical music bots, no data on that yet

the goal of @discordjs/voice is just to make a good implementation of discord's voice API in Node.js

[vocal cove]

the first example presented on github is a music bot

[vocal valley]

yes, but that does not mean I have benchmarked it against lavalink

Nor do I have any intention of doing that

I appreciate that you feel that @discordjs/voice is useless because lavalink exists,

[vocal cove]

but when you remake something, you would have goals in mind ahead, at least in theory

[vocal valley]

yep, and the goals in this remake were to:

• improve reliability
• make voice scalable (d.js V12 voice is arguably not)
• improve quality

and those have been achieved so far I would say

[vocal cove]

so it's only compared to the previous V12, maybe it will be compared to other libraries later ?

[steel crescent]

Nah, we needed to remake it either way, since we were removing built-in support.
And when you are a library of our size, relying on 3rd party for pretty much anything can get very tricky really fast (we had that like 3 times by now), where those components just ended up being unmaintained potentially. And others nor I do really have the motivation to potentially fork something like lavalink and keep maintaining it.

[vocal valley]

Potentially but that will not be done by us, our only goal is to make a good Node.js implementation of voice

You could make a similar argument of why have a Discord library in XYZ language if it already exists in another

[vocal cove]

I see what you mean,
don't get me wrong on what I've said, I really do appreciate the effort and dedication, I'm just trying to find a way to convince myself to use it instead of just using what is already existing and working, points like it will eat less ram, it will be maintained by djs already ok, hopefully more points like these arises once it's done and ready to use

[steel crescent]

I'll be brutally honest

Wait a bit still.

Wait until we have more use cases covered with examples and how to approach things easily

[vocal valley]

Yep

[steel crescent]

Especially if you have something working

[vocal valley]

If all you're looking for is stability, lavalink has been around long enough to have that claim

[vocal cove]

it won't be released by discordjs v13 ?

[steel crescent]

Yeah we will release it with v13

That doesn't mean we have 100% the best way to do things and all

[vocal valley]

if after @discordjs/voice matures and good things come of it, you can always consider switching then if you wanted

[vocal cove]

that's my issue actually, by then I would need to decide either use it and upgrade all my voice activities and the player i'm using in js, or just switch to something else

or are you saying that the old code for voice would stil work on v13 ?

[vocal valley]

What is your use case exactly? A music bot?

[vocal cove]

yea just a music bot

[vocal valley]

Old voice code will definitely not work on v13

well for a music bot the easiest and most hassle-free way is lavalink at this moment in time

[vocal cove]

i mean the music bot is a feature part of the a bot with many other features unrelated to voice

[tawdry scroll]

When will that be?

Eta

[steel crescent]

most likely by the end of this month

[vocal valley]

we'll see what people come up with using @discordjs/voice as alternatives to lavalink

[tawdry scroll]

Ight ty

[vocal valley]

but honestly if you just want voice for a music bot that you don't want to worry about, then go lavalink at this moment in time

since after all, this is new still and at least lavalink is established

[vocal cove]

I would have to remake everything either it's lavalink or the new djs voice

[steel crescent]

Whats your guild count, just curious

[vocal cove]

but I see what you mean, maybe I should just take the risk and jump in to use djs voice and hope it gets better quickly if i don't wanna remake stuff
my current voice implementation with v12 isn't that bad so far, but it will break on upgrade to 13

[vocal cove]

a dozen

[steel crescent]

like... 12?

Or a dozen as in "a lot"

[vocal valley]

oh, for small bots I would recommend @discordjs/voice

[steel crescent]

Yeah, for such a small scale you are perfectly fine here then

[vocal valley]

cause like the overhead is so much smaller compared to lavalink

[vocal cove]

it's not public yet tho, only on demand service , oh it's about 16 or so i guess

[vocal cove]

it will change once i go public tho

but maybe even at a hundred it would be still ok based on what you're saying, as a hundred is still considered small

[vocal valley]

yep

I would say so

[steel crescent]

anything < 10k is small

depending on what you do even < 50k

[vocal cove]

alrighty, I will take the risk then and bet on you then

[tall rain]

There's any docs to know anything about new voice lib?

[vocal valley]

Yep, they're linked on the repos README

I would recommend reading it alongside the examples to try and give a bit more sense to it

[vocal cove]

btw, would the next examples be on js, or ts ?

i would prefer to see js instead of ts, I know it's not much difference but still, I would need to go through converting even for just testing and play with examples

[vocal valley]

i think that's a good idea, will make the next example in JS 👍

[torn sinew]

thx will give it a try 🙂

[stoic orbit]

also whats diff then the old?

[steel crescent]

everything 🤔

[vocal cove]

thank you ❤️

[silk quartz]

pls help

[vocal cove]

member is undefined

is that coming from DM ?

[silk quartz]

no

[vocal cove]

it should be available

maybe you're using the wrong message object ?

[silk quartz]

lemme see

i am using the right one

:/

[vocal cove]

no idea why would the member be not defined in a message coming from a guild channel

[silk quartz]

voice is not defined

[vocal valley]

@silk quartz could we continue this in #archive-djs-v12-deprecated

[vocal cove]

yea true

[silk quartz]

sure

[vocal cove]

btw hydra, by using djs voice, I would still to rely on other library like ytdl ?

[vocal valley]

yep, this library is fairly low-level and so if you want youtube videos, you need to use external deps like ytdl or youtube-dl

i wouldn't be surprised to see in a few days/weeks someone makes an audio player that does this for you, though, since people have already done it for d.js v12 voice

[vocal cove]

let's say I have a link to mp3 like how you state in the example

how would djs voice stream data to discord ?

using ffmpeg ?

[vocal valley]

yep

the nice thing is that in the new voice library, we did some optimisation work with ffmpeg,

which makes it a bit faster/more efficient than v12 voice

[vocal cove]

// Attempt to convert the Track into an AudioResource (i.e. start streaming the video)
            const resource = await nextTrack.createAudioResource();
            this.audioPlayer.play(resource);

so in here where we do the streaming thing, right ?

looks like it would work a bit similar to how it was working for v12

[vocal valley]

yep 👍

in that specific example, it avoid FFmpeg because it uses ytdl-core-discord

which for most videos it can just skip the ffmpeg step and use the direct audio stream taken from youtube

[vocal cove]

yea

so it only uses ffmpeg from discord if it's a direct url to the audiofile

[vocal valley]

it uses ffmpeg in these cases:

• you pass a string to the createAudioResource function (e.g. URL, file path)
• you don't say what type of stream your audio stream is, and it cannot be inferred

[vocal cove]

oh i see

[vocal valley]

so createAudioResource(createReadStream('myfile.opus')) will still use ffmpeg,

but something like createAudioResource(createReadStream('myfile.opus'), { inputType: StreamType.OggOpus }) will skip ffmpeg

[vocal cove]

nice

I had issues with v12 about seeking, so i just started new streams with the timestamp needed

[vocal valley]

ah yes

[vocal cove]

would that be different on this verison ?

[vocal valley]

well here we just completely removed support for seeking (at least for now) because it is quite complicated to do with lots of different stream types

so if you want seeking, you can use prism-media for example to make your own FFmpeg encoder stream with the seek argument passed there

[vocal cove]

I'm using seeking just to keep the song playing from where it was if the bot changes channel

[vocal valley]

ohh

[vocal cove]

you think i should go through that just to keep track of timestamp

[vocal valley]

well actually, the library should do the work for you here anyway

when you create an audioplayer, it defaults to pausing if there is no active voice connection,

[vocal cove]

so it won't stop playing if the bot switch channel ?

[vocal valley]

when your bot switches channel, it pauses while it reconnects, then continues playing after reconnecting

[vocal cove]

that makes it better

[vocal valley]

yep, definitely a lot nicer now haha

[vocal cove]

yea i was doing that work for it on the previous version

so many workarounds

glad that things like these will be solved on the new version

[vocal valley]

this is the relevant doc for the feature if you were wondering, https://discordjs.github.io/voice/enums/nosubscriberbehavior.html

(Pause is default, i should document that)

[vocal cove]

i see

about the idle/pause

what if the bot stays idle for long

does it glitch ?

the old version if the bot stays long on the VC it won't be able to play later unless you reconnect it

[vocal valley]

so in v12 voice when we put our own time limits on things, it broke for a lot of people cause they have unstable network connections,

so now, if it disconnects for example, you can set your own limits on how long it should take to reconnect before giving up

[vocal cove]

ohhh like timeouts if if ever dc

I doubt it was connexion issue tho

[vocal valley]

we actually do it here:
https://github.com/discordjs/voice/blob/main/examples/music-bot/src/music/subscription.ts#L50

[vocal cove]

my servers never dc while it's playing

[vocal valley]

so in the music bot example, if the bot ever enters a state where it is trying to connect to discord, it's allowed 20 secs to enter the ready state before giving up and destroying

this is one part of the example i think people are safe to copy and paste since it stops them shooting themselves in the foot later

[vocal cove]

lol

what i meant is, not dc while playing

the glitch only happens when the bot is on idle for a good amount of time

like 6 hours or so

[vocal valley]

i would say that's no longer an issue in @discordjs/voice,

[vocal cove]

that's great

[vocal valley]

it might have been in v12 since basically we didn't check whether we could send audio packets to discord (after some time, it ends up closing off)

[steel crescent]

youd be surprised how fragile voice actually is, especially with the node event loop. if you run heavy computations that could potentially lock up or clog the event loop, it will drastically affect your music too
and newcomers usually have no idea whats expensive and what not, so those pitfalls get even bigger imo.

[vocal valley]

but in /voice we do like heartbeats now, where discord echos back the data, if it doesn't echo back after 1 minute, the connection counts as disconnected

[vocal valley]

yep, hopefully this will be eased up by being able to run it in a childprocess or something now

[vocal cove]

that's the main reason I have all computations in another server

my voice server never lag or have anything to do beside doing the audio streaming

[vocal valley]

a very good idea lol

[vocal cove]

well, early i did separate only heavy computations about generating graphics and such, but later i just migrated all to be separated to the voice server

I will try out the new djs voice soon, maybe once I start upgrading everything to v13

[vocal valley]

glad to hear 😄 hopefully by then the nicer high-level doc about the library will be ready too

and maybe even audio player libraries that use @discordjs/voice

[vocal cove]

I m using my own player even in v12, cuz I do use many customizations either on visuals or the way how it works

I will prolly be doing the same on v13

[vocal valley]

fair

[silk quartz]

hey

about that issue befor

if i change my discord.js to v13

would it solve my problem

?

[vocal valley]

nope, that's an issue with your d.js config itself

[silk quartz]

oh ok

[vocal valley]

https://discordjs.guide/popular-topics/intents.html might be useful to u

[silk quartz]

thanks

[frosty quartz]

Thanks 😄 !!! ❤️

[red nest]

There is so many new feature in v13

[pliant basin]

what were the results?

[carmine timber]

@pliant basin These were the results

[pliant basin]

oh i thought you meant you tested the performance of lavalink against @discordjs/voice because the person you were replying to had their message right below a message about the benchmark haha 😅

still, thanks for sharing the results for @discordjs/voice!

[carmine timber]

😂

[frosty dawn]

Error: Did not enter state ready within 60000ms Bot is not entering ready state

    const connection = joinVoiceChannel({
        channelId: channel.id,
        guildId: channel.guild.id,
        adapterCreator: channel.guild.voiceAdapterCreator,
    });

    try {
        await entersState(connection, VoiceConnectionStatus.Ready, 60e3);
        return connection
    } catch (error) {
        connection.destroy();
        throw error;
    }
}```

here whats i am trying to do its for a stage channel

the bot is entering the channel but throws a error not ready

Can anyone provide a solution?

[icy maple]

Make sure you have the GUILD_VOICE_STATES intent enabled

[frosty dawn]

thanks man solved it

[subtle granite]

We’re certain music bot libraries affected with v13? As such, is the new discord voice stuff required? I used to use Lavalink which worked in v12 and and upon switching to v13 and getting things up to date, it fails to transmit any voice or speak.

[fervent estuary]

lavalink should work with minimal changes. This effects everyone using v12 built in voice

[subtle granite]

Huh. Maybe I’m doing something wrong then. It just totally fails to play with any of the libraries I’m using

[carmine timber]

Show me code

[subtle granite]

            guild: interaction.guild.id,
            channel: channel.id,
            node: "1"
        });

if (!player)
  return;

await player.play(track);```

I mean its pretty basic code. Nothing crazy, I am using Lavacord, I create the player, it joins the channel and attempts to play the track but no data is transmitted. It used to work like this in v12, but broke in v13. I was just curious if something glaring changed and I needed to adjust or not.

[slender sparrow]

no need to change if you use lavalink

[carmine timber]

Lavacord is different

[carmine timber]

I prefer you to switch to LavaLink

[slender sparrow]

he should change from lavacord to other client

lavacord not maintained for a long time

[carmine timber]

Yes

[subtle granite]

Any recommended other clients? I’ve seen a few.

[slender sparrow]

shoukaku, erela.js

https://github.com/freyacodes/Lavalink#client-libraries

[carmine timber]

LavaLink, erela.js

[subtle granite]

Alright I’ll give them a shot.

Thanks for the help

[merry igloo]

If I have a file like an .mp4, will I be able to do player.play(musicfile.mp4)?

Assuming everything else like the player and connection is already set up

[carmine timber]

Yes ofcourse

[carmine timber]

But you need to subscribe the connection with player

[merry igloo]

Thank you

[carmine timber]

Sorry, You need to create Audio Source first for that and then include that source in player

[merry igloo]

Oh ok, thanks again

[carmine timber]

Like this : https://deploy-preview-595--discordjs-guide.netlify.app/voice/audio-player.html#playing-audio

[gritty pasture]

just a question, can you use js to code discord.js@voice cuz the library is made with ts

[stark drum]

Of course you can, when published to npm code is being compiled into JavaScript
Compiled code is available in the dist/ folder

[vocal valley]

will also be making some examples in JS for those who haven't used TS before

[carmine timber]

How do you compile a ts file ??

[stark drum]

I've got a question about how the library is designed
Is it possible, or would it be possible, by its nature to use it to make a separate voice server? Something like Lavalink but a bit more customizable?

[vocal valley]

Using tsc,

The music bot example actually has typescript compilation, you could take a look like that

[vocal valley]

I would say so yes, this library can be seen as the low-level server you need to interact with the voice API

[carmine timber]

But tsc convert ts to js only, how you convert it to .d.ts and .d.ts.map ??

[vocal valley]

So you could host this somewhere else separate to your bot, like lavalink

[vocal valley]

You need to enable the declaration properties in your tsconfig, if you have a look at the voice repo's tsconfig, we have our config set up to output these files too

[carmine timber]

Ok

Thanks

[stark drum]

Oh great, looks promising 😄
Congrats to the team btw!

[vocal valley]

Np

[gritty pasture]

ya, the only reason i asked the ques was because the examples were in ts , so ya also thanks

[vocal valley]

np

[carmine timber]

@vocal valley Can we use new voice API with lavalink ??

[vocal valley]

This library isn't really related to lavalink

you can still use lavalink with discord.js v13 though

[carmine timber]

I know that

[vocal valley]

but you can't use lavalink with @discordjs/voice, since lavalink uses lavaplayer as its backend

[carmine timber]

Ok I got it

[carmine timber]

Will new Voice API release on v13 launch date?? Since it is very close and moreover, proper docs are not available still yet.

[steel crescent]

Yes

[carmine timber]

Oh, Then all music bot developers on v12 voice needs to update to this soon 😂

[steel crescent]

Yep 👍

[subtle granite]

This sounds interesting

quick question

do u allow opus encoder args or not

if not then do it then your module / addition to v13 is AMAZING

Because the Docs are so weird to read, you need to click everything and go back and forth all the time I won't spend the time to read it through everything

And if you allow spawning ffmpeg things, then do u do it with child_process? if not, do it with it, otherwise lags and slow-

[carmine timber]

yes they do allow to stream in opus

[carmine timber]

Are you talking about this : https://deploy-preview-595--discordjs-guide.netlify.app/voice/ ??

[sacred cove]

when support will end for the current voice system in v12, will it still be usable or will we have to immediately switch to the new library? Because I think most people (including me) need some time to update their code for the new library, which is almost finished, but still not yet, so I doubt people will update at this time

[carmine timber]

Switch to new library as launch of v13 is very close and it will replace v12 voice api.

[sacred cove]

but the library isn't finished yet, it might be that they change something and then I'd need to update again

[carmine timber]

No library is almost complete, docs is incomplete

[sacred cove]

obviously I don't mean that it's incomplete the same way as djs is, just saying that it's a brand new library, and as they said it's still in development no wait didn't get what you said in the first place lol

yes, the library is indeed almost complete

but also without the docs it's not that easy to update

[carmine timber]

What do you need for updating code ?? I will give you code

[rustic temple]

Btw, it's still installing, is it ok? XD

[subtle granite]

Do i still need @discordjs/opus for this to work?

[sacred cove]

I'm not updating yet, I just wanted to know if it's necessary to update right now, since it'll be much work for me

[carmine timber]

no

it is optional

[subtle granite]

I am talking about the official Docs: https://discordjs.github.io/voice/classes/audioplayer.html

[carmine timber]

Then All the best

[echo bone]

I don't unterstand how i can join a vc with the new library...

Err:

(node:27024) UnhandledPromiseRejectionWarning: TypeError: adapterCreator is not a function

Code:

const MemberVoiceChannel = Interaction.member.voice.channel;
const VoiceSubscription = await DiscordVoice.joinVoiceChannel(MemberVoiceChannel);

[carmine timber]

You need to specify adapter creator

it is mandatory in v13

[echo bone]

And how can I do it?

[carmine timber]

Wait let me show u

[echo bone]

Thanks

[carmine timber]

var connection = getVoiceConnection(message.guild.id)
  if (connection?.state.status !== VoiceConnectionStatus.Destroyed){
    connection =  joinVoiceChannel({
        channelId: message.member.voice.channel.id,
        guildId: message.guild.id,
        adapterCreator : message.guild.voiceAdapterCreator,
        selfDeaf : false,
        selfMute : false
  })}

[carmine timber]

For guild.voiceAdapterCreator you need latest @dev build

[echo bone]

@carmine timber Yeeesss, thanks. It works now... ❤️

[carmine timber]

❤️

[echo bone]

I don't unterstand the err... Got IT

[echo bone]

I found the first error, but now I got another error by @discordjs/voice

TypeError: secretbox.methods.close is not a function
  at Networking.encryptOpusPacket (/home/disgroup/DisBot/node_modules/@discordjs/voice/dist/networking/Networking.js:413:35)
    at Networking.createAudioPacket (/home/disgroup/DisBot/node_modules/@discordjs/voice/dist/networking/Networking.js:397:53)
    at Networking.prepareAudioPacket (/home/disgroup/DisBot/node_modules/@discordjs/voice/dist/networking/Networking.js:321:37)
    at VoiceConnection.prepareAudioPacket (/home/disgroup/DisBot/node_modules/@discordjs/voice/dist/VoiceConnection.js:282:33)
    at /home/disgroup/DisBot/node_modules/@discordjs/voice/dist/audio/AudioPlayer.js:419:54
    at Array.forEach (<anonymous>)
    at AudioPlayer._preparePacket (/home/disgroup/DisBot/node_modules/@discordjs/voice/dist/audio/AudioPlayer.js:419:19)
    at AudioPlayer._stepPrepare (/home/disgroup/DisBot/node_modules/@discordjs/voice/dist/audio/AudioPlayer.js:391:22)
    at prepareNextAudioFrame (/home/disgroup/DisBot/node_modules/@discordjs/voice/dist/DataStore.js:71:31)
    at audioCycleStep (/home/disgroup/DisBot/node_modules/@discordjs/voice/dist/DataStore.js:56:5)

And this keeps crashing my whole node process

[subtle granite]

Anyone use erelajs with v13? Got recommended to use it. Switched my project to it, it connects to the node, and attempts to the play the track, but nothing is transmitted in the channel. (works fine on v12)

Seems that there is no sessionId or anything like it doesnt attempt to do voiceUpdate with Discord at all.

[vocal valley]

can you install tweetnacl or libsodium-wrappers or sodium?

[vocal valley]

You don't have to immediately update once v13 releases, you can take your time 👍

[sacred cove]

that'll be easier then, good to know

[echo bone]

Which one do you think is the best?

[vocal valley]

Sodium is best, then libsodium-wrappers, then tweetnacl

Sodium can be quite a pain to install though

[fiery blade]

are there types for this lib?

[carmine timber]

erelajs is a third party plugin, so no is going to help here, ask erelajs author about this.

[slender sparrow]

why instant finished?

[carmine timber]

show me code

[slender sparrow]

nvm fixed, i was using discord-ytdl-core instead ytdl-core-discord, looks like @discordjs/voice didnt support discord-ytdl-core

[carmine timber]

Oh

[carmine timber]

BTW new voice API ??

[slender sparrow]

wdym, @discordjs/voice ?

[carmine timber]

Oh nvm

[subtle granite]

how do i leave a voice channel

[carmine timber]

const { getVoiceConnection } = require('@discordjs/voice');

const connection = getVoiceConnection(myVoiceChannel.guild.id);
connection.destroy();

[subtle granite]

oh okay

thanks :>

[vocal valley]

Yes they are included with the install

[vocal valley]

To be honest it should support both

[fiery blade]

Do you know what causes this then? I definitely have the module installed

[vocal valley]

Ah, you need to enable skipLibCheck in your tsconfig

[fiery blade]

oh okay

[subtle granite]

im also experiencing this problem

[carmine timber]

hydrabolt will fix this soon

[vocal valley]

yep, it is fixed on GitHub, will be included in next release

Basically the dependency report doesn't look far enough for each modules package.json file, which is why it couldn't find a version ^

now fixed though 👍

[subtle granite]

oh okay

[boreal solstice]

would this ever support video streaming with bots 😔

[vocal valley]

nope, discord isn't documenting it since it is too complicated for bots

[grim fulcrum]

Hi, how I change a playback volume?
resource.volume = 0.5 doesn’t work.

[vocal valley]

@grim fulcrum did you enable inline volume?

const resource = createAudioResource(uri, { inlineVolume: true });
resource.volume.setVolume(0.5);

that should work

[grim fulcrum]

Thanks, it works.

[vocal valley]

np 😄

[vernal ibex]

Hi, everyone. I'm new to this server, nice to meet you all. Had a question about Discord API - I know with the voice API you can send voice messages. Is there anyway that I can subscribe to other users' voice stream (for purposes like transcribing)? with djs or not with djs. Thx

[vocal valley]

@vernal ibex yep this is possible!

[dusty needle]

VoiceReceiver
Receives audio packets from a voice connection.

[vocal valley]

you can subscribe to users with this

specifically for our new voice library @discordjs/voice, voice receive is a little rough around the edges still, but should be improved soon

[vernal ibex]

Amazing. Thanks a lot @vocal valley

[carmine timber]

Do sodium requires VS 2010 - 2015 windows build tools ??

[steel crescent]

2015/2017/2019

but good luck on windows

oh

its outdated af

good luck

[carmine timber]

Yes

[steel crescent]

use libsodium-wrappers or so

fuck sodium

esp on windows

[carmine timber]

XD

[carmine timber]

BTW, why are you recommending outdated things in new voice API ??

[steel crescent]

?

its basically sodium but in pure JS

nothing about it is outdated

[carmine timber]

But We can't compile it

That's what matters

[steel crescent]

What?

libsodium-wrappers is webassembly

[carmine timber]

I am unable to install npm i sodium

[steel crescent]

yes don't use that

[carmine timber]

Ok

[steel crescent]

use it on linux or so

but not when on windows

[carmine timber]

Ok got it

[honest shuttle]

So what has better performance? libsodium-wrappers or sodium?

Or tweetnacl maybe?

[rustic temple]

Iirc use sodium in Linux and libsodium in windows (idk actually)

[steel crescent]

Doesn’t matter

Tweetnacl performs the worst though

[mossy flower]

is this the end of lavalink? 👀

[next thicket]

How to detect which voice channel I'm in? In v13?

[mossy flower]

<Guild>.voice.channel I think (yeah i am pretty sure thats it)

[next thicket]

need to fix, ¯_(ツ)_/¯

[steel crescent]

Is this actually concerning the new voice module

[next thicket]

Yea ik

[steel crescent]

Or just old v12 voice

[next thicket]

This?

[steel crescent]

Yeah that goes to #archive-djs-v12-voice-deprecated if you aren’t on the master branch or v13

[next thicket]

console.log (song.url);
const dispatcher = queue. connection
.play(ytdl(song.url, { filter:
"'audioonly" }))
``` then I failed suffering v13

[steel crescent]

V13 is very different from that if you check the pins

[subtle granite]

how do you specify bitrate?

[silver sleet]

[vocal valley]

at the moment:

resource.encoder?.setBitrate(...);

[jolly mason]

Pls help
TypeError: adapterCreator is not a function

const connection = joinVoiceChannel({
    channelId: message.member.voice.channel.id,
    guildId: message.channel.guild.id,
    adapterCreator: message.guild.voiceAdapterCreator,
})```

[vocal valley]

are you using dev branch of d.js?

[vocal valley]

you need to use a more recent of dev

[jolly mason]

oh ok

[jolly mason]

is there any example that shows playing audio from ytdl-core to a voice channel using new library in javascript?

const connection = joinVoiceChannel({
    channelId: message.member.voice.channel.id,
    guildId: message.channel.guild.id,
    adapterCreator: message.guild.voiceAdapterCreator,
})

const player = createAudioPlayer()
let resource = createAudioResource(await ytdl(ytLink))
await player.play(resource)```

Bot isn't playing any audio  after updating to new library code and showing no errors, pls help 

Debugging libraries:
```prolog
Core Dependencies
- @discordjs/voice: 0.3.1
- prism-media: 1.2.9

Opus Libraries
- @discordjs/opus: 0.5.3
- opusscript: not found

Encryption Libraries
- sodium: not found
- libsodium-wrappers: 0.7.9
- tweetnacl: not found

FFmpeg
- version: 4.3.2-2021-02-27-essentials_build-www.gyan.dev
- libopus: yes```

?

[vocal valley]

@jolly mason you need to subscribe your connection to the player

we have a typescript example with YouTube, and if you remove all the type annotations and interfaces you see, it becomes JavaScript

there will be some actual JS examples too however

[south lark]

Is receiving voice from users on web browsers causing the bot to completely crash still an issue with this new voice stuff?

[analog tundra]

Can i use this Voice Library with Lavalink and Erela?

ignore this ^^

[carmine timber]

Show me code ??

[carmine timber]

How to change volume of playing song in new voice api ??

[vocal valley]

I tried yesterday and couldn't replicate the error, however it was also working fine on v12 voice yesterday for me too

[vocal valley]

#archive-voice message

[carmine timber]

Thnx

[vocal valley]

If you are still able to replicate this on v12, would you ping me or something? 👀

I need to collect some data to fix the problem, and it's a bit hard since I'm unable to replicate

[carmine timber]

Is playbackDuration excluding pauses or including pauses ??

[vocal valley]

interesting question let me double check 👀

[vocal valley]

So audio player's playbackDuration includes pauses, whereas the resource's excludes the pauses

[carmine timber]

Audio Player include PlaybackDuration ?? https://discordjs.github.io/voice/classes/audioplayer.html

[vocal valley]

Yep, if it is in the playing, paused, or autopaused states

https://discordjs.github.io/voice/interfaces/audioplayerplayingstate.html

[carmine timber]

Ok

[carmine timber]

It is completely different as compared to v12

See pins

[icy maple]

Just read the guide

[carmine timber]

Nothing got removed, instead whole new library was created in a new way

You no longer can do <GuildMember>.voice.channel.join() in this

[split cliff]

What are the new features for voice if 13.0.0 cones out?

[carmine timber]

See pins

First you need to join a channel and then play something

[wise anchor]

joinVoiceChannel().play doesn't work

[wise anchor]

yes i already did the join channel part

[carmine timber]

It would not

[carmine timber]

Show me

code

[carmine timber]

it connects ??

[wise anchor]

yes it joins the vc

[carmine timber]

For playing you need to something like :

            var source = createAudioResource(<ytdl stream>)
            var player = createAudioPlayer()
            player.play(source)
            connection.subscribe(player) // For you connection = queue.connection

[wise anchor]

also another question

[carmine timber]

ask

If you don't update your djs

Then possible 🙂

[wise anchor]

oh

now i have to change a lot of code

[carmine timber]

Yes

[carmine timber]

Do you need code with in line volume controller ??

[wise anchor]

wdym

[carmine timber]

Volume control in player ??

[wise anchor]

yes i need to be able to change the volume of the video

[carmine timber]

So do this :

var source = createAudioResource(<ytdl stream>, { inlineVolume : true })
            var player = createAudioPlayer()
            player.play(source)
            connection.subscribe(player) // For you connection = queue.connection

Then only you will be able to use volume controller

[split cliff]

Ah, ty!

[subtle granite]

Do I still need to install @discordjs/opus and ffmpeg?

[carmine timber]

Read This : https://deploy-preview-595--discordjs-guide.netlify.app/voice/#installation

[subtle granite]

Oh did not know this new guide exists

[carmine timber]

You could have seen pins

[carmine timber]

Code ??

Where is .set ??

[wise anchor]

nowhere

[carmine timber]

Can you show me full error

??

[wise anchor]

sure

[vocal valley]

what version of d.js do you have installed?

you'll need one of the latest dev versions (as in last day or so)

[wise anchor]

i have the latest

[carmine timber]

Type npm list discord.js

[vocal valley]

can you log your guild object for me?

maybe it is a partial guild

[wise anchor]

ight

[carmine timber]

Why are you extending channel ID so long

??

[wise anchor]

good question

[carmine timber]

@vocal valley , he has some mistakes in joinVoiceChannel

Nothing else

[distant osprey]

client.guilds.cache.get().channels.get().guild.xxx

[vocal valley]

the joinVoiceChannel looks a bit verbose, but i think it's correct

[rustic temple]

They're string, not number

[wise anchor]

ight

same error

[vocal valley]

can you console.log the guild you get

[carmine timber]

What is your 577 line ??

[wise anchor]

i don't have a 577 line

[carmine timber]

Sorry

[wise anchor]

i have 29 lines

[vocal valley]

seems to be failing on this line in d.js:

this.client.voice.adapters.set(this.id, methods);

which is odd

[carmine timber]

Can you log everything after adapterCreator??

[vocal valley]

hmm ok, this might be an issue in d.js dev, i'll take a look now

there was a PR that removed unused voice code, maybe it accidentally touched some of the new adapter code

cannot replicate on latest dev release 🤔

can you also log guild.client.voice

[wise anchor]

okay

[wise anchor]

do i log it before or after the join part?

[vocal valley]

before should work i think

[vocal valley]

🤔

how on earth

can you uninstall discord.js and try reinstalling it

[wise anchor]

okay

[vocal valley]

im not really sure how this has happened, but that is v12 voice code

this is so cracked umm

i think your node_modules might be a bit messed up

or your npm install

[wise anchor]

and how do i fix

[vocal valley]

i have no clue tbh 😬 maybe reinstall nodejs?

[carmine timber]

Just make a new folder and shift your main files to that folder

[wise anchor]

okay

[carmine timber]

and then install modules separately

[hazy copper]

i keep getting this error most of the time on playback using this AudioPlayerError: aborted

[carmine timber]

Code ??

[hazy copper]

error gets emitted by the error event

its just a stream being played over a player

const connection = joinVoiceChannel({
    channelId: interaction.member.voice.channel.id,
    guildId: interaction.guild.id,
    adapterCreator: interaction.guild.voiceAdapterCreator,
    //selfDeaf: false
});

const ytdlstream = ytdl(url, { filter: 'audioonly' });
const stream = createAudioResource(ytdlstream);


const player = createAudioPlayer({
    behaviors: {
        noSubscriber: NoSubscriberBehavior.Pause
    }
});
player.play(stream);

const subscription = connection.subscribe(player);

[wise anchor]

same error but i fixed the corruption issue

[hazy copper]

just testing this new module before replacing my old one

[carmine timber]

Can you change this const player = createAudioPlayer(); ??

and then tell if error persists or not

[hazy copper]

you mean remove the object from it?

[carmine timber]

yes

just for testing purposes

[hazy copper]

aight will leave it playing for a minute to see if i get it

[carmine timber]

Ok

[carmine timber]

did you removed the code ??

[wise anchor]

Its still the same code but i reinstalled discord.js

[carmine timber]

I needed the codes

to check any issues

[wise anchor]

Okay

So now what do i do?

[carmine timber]

Show me once again

[wise anchor]

Do i run the command

Again

Okay i will do that but im not at home atm

[vocal valley]

@hazy copper can you try the example on this branch and tell me if you can still repro? https://github.com/discordjs/voice/tree/feat/probing

it uses youtube-dl instead of ytdl-core

[carmine timber]

Page Not Found for me @vocal valley

WTF

[vocal valley]

oops

wrong one

try now

[hazy copper]

yeah i was thinking of moving to that as well, it returns a webm stream by default right? avoiding ffmpeg convertions

[vocal valley]

yep 😄

the only difference is that youtube-dl downloads the video rather than ytdl-core

so far ive not experienced any streams that end early with youtube-dl

[hazy copper]

where is the example exactly? the one i found uses ytdl-core?

[vocal valley]

it will use ytdl-core but only to get the video info

https://github.com/discordjs/voice/tree/feat/probing/examples/music-bot

this uses ytdl-core to get video info, but youtube-dl to actually stream the video

code for it is here https://github.com/discordjs/voice/blob/feat/probing/examples/music-bot/src/music/track.ts#L46

[hazy copper]

will try this rn

never done typescript before so give a sec to my slow brain lul

[vocal valley]

np!

just try to ignore the type annotations

[carmine timber]

I am getting error @discord/voice has no exported member 'demuxProbe'

What to do ??

[vocal valley]

Can you rebuild the /voice files

so npm run build in the /voice dir, then cd into the example and try again

[carmine timber]

Oh so long process

[vocal valley]

not too long 😆

[hazy copper]

hmm i cant run this example bot

saying it cant find @discordjs/voice but i just did an npm i

[vocal valley]

might need to reinstall discord.js@dev

and then also npm run build on the root repo

[hazy copper]

aight just did the build

yep it printed ready

[carmine timber]

How to run a ts file ??

node <file>.ts ??

[hazy copper]

getting this whenever i try to play something

[carmine timber]

How did you run a ts file ??

[hazy copper]

just use the start script

npm start

[carmine timber]

ok

[vocal valley]

can you npm I tweetnacl

[hazy copper]

yep that was it, song is playing now

[vocal valley]

amazing 😁

let me know if you run into any errors

[hazy copper]

i will leave this running for 5mins

that was how long i was running it to get the error on the other one

[vocal valley]

👍

[hazy copper]

are there any advantages when using TS over JS?

[vocal valley]

it's mainly just avoiding making type errors imo

you can still get the intellisense of @discordjs/voice when you use JS

But it's easier to be confident in the code when you know it has been type checked too

[hazy copper]

i see, haven't gotten any errors yet

[vocal valley]

😁

[hazy copper]

wouldn't this use more bandwidth?

[vocal valley]

I don't think it would?

[hazy copper]

well as soon as i spoke

got the same error

[vocal valley]

Can you send a screenshot of the error in your console

👀

[vocal valley]

just to be sure, can you send me your dist/music/track.js file

[hazy copper]

could it be my node version by any chance? im on the latest one

yep i got you fam give me a sec

[vocal valley]

oh, I meant like the compiled .js file

[hazy copper]

oh ye my bad haha

[vocal valley]

Ahh seems like it hasn't been recompiled

are you definitely on the right branch?

[hazy copper]

its the one you sent me idk

[vocal valley]

if you are, could you try deleting the dist folder inside examples and trying again

[hazy copper]

try building again?

[vocal valley]

try deleting dist and recompiling, in that same file, it should import ytdl-core, not ytdl-core-discord

yep

And branch should be feat/probing

[hazy copper]

i will redownload the branch to make sure

that's correct right?

[vocal valley]

I think so,

[hazy copper]

oh yep i was on the wrong branch

[vocal valley]

you can double check if it's right by going to src/utils, there should be a demuxProbe.ts file

ahhh

[hazy copper]

download folder

haha

[vocal valley]

Lol

[hazy copper]

sorry for wasting your time on this lulw

[vocal valley]

it's ok np

[hazy copper]

i got to npm i everything again 1sec

[vocal valley]

yeah np

It is a pain, I've had the same annoyances with it when making it

I think I messed up the tooling somewhere

[carmine timber]

I am getting this error Error: Cannot find module 'tsconfig-paths/register'

What to do ??

[vocal valley]

can you try npm i again?

[carmine timber]

Ok

Now it worked 😄

[vocal valley]

😁 nice

[hazy copper]

is that because youtube-dl isn't installed?

[vocal valley]

yep I think so

the README has installation instructions if you have python and pip

Otherwise there's a link where you can install it elsewhere

[hazy copper]

yep 1 second my python install is scuffed i think

alright fixed it now

[vocal valley]

👍

Is it working now?

[hazy copper]

song is playing now, will leave it running

[vocal valley]

Nice 😁

Fingers crossed

ping me if it breaks or anything, or if it works

[hazy copper]

yep, i just hope it doesn't use much bandwidth because im self hosting this on a raspberry pi

[vocal valley]

I think it should be fine, but unsure

Ooh you can apparently set a download speed limit on the song

If you want a guarantee on bandwidth cap, you could try that

[hazy copper]

sounds good, i guess this would use less bandwidth overall since it downloads it and then it doesn't download anything for the rest of the stream

[grim fulcrum]

Example bot in feat/probing branch ends playback without starting it, with no any errors.

[vocal valley]

👀

can you try the very latest commit

and then run this:

npm i && rm -rf dist && npm run build && cd examples/music-bot && rm -rf dist && npm i && npm start

and also double check that youtube-dl is installed

if it still does the same thing after that, can you ping me with your node.js version and os

[hazy copper]

@vocal valley have been running it since my last message, looks like this doesn't get any errors

[vocal valley]

amazing! 🎉

very happy to hear that

[vocal valley]

can you send node_modules/src/structures/Guild.js

[wise anchor]

okay

[vocal valley]

oh actually not that

send node_modules/discord.js/src/client/voice/ClientVoiceManager.js

[vocal valley]

yep, that's the outdated voice file

[wise anchor]

and what do i do now?

[vocal valley]

can you cd to the Workroom folder

then npm i discord.js@13.0.0-dev.02693bc02f45980d8165820a103220f0027b96b7

[wise anchor]

okay

[vocal valley]

oops

try now 😅

[wise anchor]

done

[vocal valley]

i am confused why only one of the files seems to be outdated

ok, can you send the new file at that location

[wise anchor]

okay

[vocal valley]

what the heck

[hazy copper]

@vocal valley you have any idea why ytdl-core is having these issues?

[vocal valley]

honestly nope, it's had them since people started making youtube bots with it :/

[wise anchor]

i am 100% sure that is the correct file

[vocal valley]

im guessing some faulty reconnect logic

[vocal valley]

delete your node_modules folder,

[wise anchor]

okay

[vocal valley]

make sure in package.json discord.js is pinned to 13.0.0-dev.02693bc02f45980d8165820a103220f0027b96b7

then try installing all your modules again

[wise anchor]

okay

[vocal valley]

also delete a lockfile if you have it

maybe that's screwing stuff

[vocal valley]

yep that's fine

[wise anchor]

okay done

i deleted node modules and package-lock.json file

[vocal valley]

ok, now if you reinstall, hopefully the file should be different and a lot smaller

[wise anchor]

what do i run?

npm install?

[vocal valley]

yep

[wise anchor]

ight

[steel crescent]

just always run npm i --force

it will refetch

[grim fulcrum]

Yes, the error persists.
Node.js version: v16.0.0 (I also tried to run on v16.3.0)
OS: Ubuntu 20.04.2 LTS

[wise anchor]

what the bruh

i am getting this again...

[vocal valley]

youtube-dl -o - -q -f "bestaudio[ext=webm+acodec=opus+asr=48000]/bestaudio" -r 100K https://www.youtube.com/watch?v=VBKNoLcj8jA > out.webm

can you send me the out.webm file you get after running this in a terminal

and also can you send the video URL that is causing you the issue of the bot starting and stopping immediately

[grim fulcrum]

I'm sorry, the problem is on my side.

ERROR: VBKNoLcj8jA: YouTube said: Unable to extract video data

[vocal valley]

oh huh

are you in a georestricted region maybe?

[steel crescent]

just pick a random video yourself instead of taking the one hydra gave you

[grim fulcrum]

It doesn't work for me.

[steel crescent]

What doesn't work?

And what video did you try?

[grim fulcrum]

I do not know, I have never used this library. I will try to solve this problem.

[wise anchor]

ight i fixed the corruption issue

[grim fulcrum]

I tried several videos, but still the same error occurs.

[vocal valley]

nice! should work now 😄

[wise anchor]

ight thx for the help

[vocal valley]

np

[wise anchor]

okay so it joins the channel but

and it doesn't play anything

[vocal valley]

try npm install libsodium-wrappers

and make sure you're using the latest commit on that branch

[wise anchor]

thx

[vocal valley]

np 😄

[wise anchor]

a few questions

so how do i get the "dispatcher"

for example in old system you could do

let dispatcher = connection.play(...)

and then you could set the volume and stuff

how do i do that?

[vocal valley]

so volume:

const resource = createAudioResource('thing.mp3', { inlineVolume: true });
resource.volume.setVolume(0.5);

as for the "dispatcher", the concept of a dispatcher is gone and it's more of just a player (AudioPlayer) and a resource (AudioResource)

i think it'd be good to look at both the music bot example and this doc https://deploy-preview-595--discordjs-guide.netlify.app/voice/

to move onto the next song:

player.on('stateChange', (oldState, newState) => {
  if (newState.status === AudioPlayerStatus.Idle) {
    player.play(getNextResourceFromSomewhere());
  }
});

whenever an audio player enters the idle state, it means it has finished playing something

if you want i think it's also good to copy the state transition logic in the music bot example since it is known to work

[wise anchor]

ight

[grim fulcrum]

The problem with youtube-dl is solved. The bot from the example works.

[vocal valley]

great 😄

how did you fix your issue with youtube-dl out of curiosity?

[wise anchor]

how do i pause a song? what about resuming?

[grim fulcrum]

I just reinstalled pip and youtube-dl 😅

[vocal valley]

ah fair

[vocal valley]

player.pause()
player.unpause()

[carmine timber]

I need you in #archive-djs-v12-voice-deprecated

Come there

[wise anchor]

what about seeking

[vocal valley]

atm it doesn't support seeking since it was half broken on v12 anyway, you can enable it yourself like this:

const prism = require('prism-media');
const { createAudioResource, StreamType } = require('@discordjs/voice');

// These are arguments used to convert the input to a format suitable for @discordjs/voice
const FFMPEG_ARGUMENTS = [
    '-analyzeduration', '0',
    '-loglevel', '0',
    '-f', 's16le',
    '-ar', '48000',
    '-ac', '2',
];

// An audio stream starting from 10mins onwards in myfile.mp3
const stream = new prism.FFmpeg({
    args: ['-ss', '00:10:00', '-i', 'myfile.mp3', ...FFMPEG_ARGUMENTS],
});

const resource = createAudioResource(stream, { inputType: StreamType.Raw });

[wise anchor]

and another question

i used to do queue.dispatcher.destroy()

how do i do that now

[vocal valley]

if you want to stop playing a song on an audio player, player.stop()

[carmine timber]

You can do connection.destroy() to disconnect and player.stop() for stopping player

[wise anchor]

ight

there was also channel property to dispatcher

how can i get that now

[vocal valley]

connection.joinConfig.channelId

[wise anchor]

ight ty

i am now getting this queue.connection.channel.leave() is not a function

[carmine timber]

No

[wise anchor]

this used to be

[carmine timber]

@wise anchor Use this

[wise anchor]

okay ty

[lost lynx]

Can I 'simply' use this module with discordjs v12?

[carmine timber]

which module ??

[lost lynx]

@discordjs/voice

[carmine timber]

Idk

@vocal valley

[vocal valley]

yes, i would say so

[carmine timber]

but he will require a adapter

[vocal valley]

examples/basic is designed to work with v12

the only difference is adapter, yep

[carmine timber]

file also

[lost lynx]

I want to see if the error I get still occurs if I migrate my simple code to this lib. First steps first, I get a "adapterCreator is not a function" from new VoiceConnection so...

[vocal valley]

so, you can use this file: https://github.com/amishshah/discord-radio-bot/blob/feat/upgrade/src/adapter/adapter.js

[carmine timber]

😄

[vocal valley]

and instead of guild.voiceAdapterCreator, it'd be createDiscordJSAdapter(channel)

[wise anchor]

now i have to remake the pause and stop and resume

[carmine timber]

👍

[vocal valley]

after this the rest should be identical, only difference is in adapter

[lost lynx]

Thanks, I'll try that.

[vocal valley]

np

[wise anchor]

and the seek

oh and is this still a thing?

[carmine timber]

Yes but with resourse.playbackDuration

https://discordjs.github.io/voice/classes/audioresource.html#playbackduration

[wise anchor]

so i need to get it from the AudioResource and not from the connection?

[carmine timber]

yes

[vocal valley]

you can either get it from the player or audio resource, the one from the player includes any silence packets that are played

resource is probably better for your use case

[wise anchor]

ight

adapterCreator is not a function

[carmine timber]

Are you using it in message event or interaction event ??

[wise anchor]

i am using it in a command

[carmine timber]

what type of command ?? slash or normal ??

[wise anchor]

normal

[carmine timber]

djs version ??

[wise anchor]

[e-mail]14099779e61cb84dcd1cb2497e0e853a6144