#Help with configuration
1 messages ยท Page 1 of 1 (latest)
there have been quite a few changes to various things recently and alot of videos are now out of date (even those only a few months old)
what do you have do far?
can you share your esphome config?
is it failing to build or just not working once built?
do you have it wired up using one i2s bus or two?
i have connected it using i2s bus
i2s_audio
its definetly i2s ๐
i meant i2s**
your not using a media player component in the chain which is what would normally be used
i have a mostly working config thats very simalar give me a moment ill adapt it for your setup. you might need to adjust some but hoepfully will get you to a working state and can maybe go from there
ok thanks
whats with all the specific options set here?
board: esp32-s3-devkitc-1
framework:
type: esp-idf
sdkconfig_options:
CONFIG_ESP32S3_DEFAULT_CPU_FREQ_240: "y"
CONFIG_ESP32S3_DATA_CACHE_64KB: "y"
CONFIG_ESP32S3_DATA_CACHE_LINE_64B: "y"
CONFIG_AUDIO_BOARD_CUSTOM: "y"
psram:
mode: octal
speed: 80MHz```
I just saw a guide by https://www.youtube.com/watch?v=rT_q0g8dtmY from this guy which kinda worked so i kept using his config
Let's look at how we can setup On-Device Wake Word Detection for your Voice Assistant in Home Assistant using ESP32 S3.
Setup Guide & Circuit Diagram : https://smarthomecircle.com/How-to-setup-on-device-wake-word-for-voice-assistant-home-assistant
===========
Components used in the video
ESP32 S3 N8R2 :
AliExpress ...
i think it is supposed to be like some performance thing
correct
can give that code a try
yes i am compiling it
it doesnt have mute and only has single wakeword loaded
this is my little project speaker ๐
oh nice
old pc speaker i found in a box that i started messing with. although for the most part i just use voice-pe's
my speaker is a mmwave sensor with voice assitant built in
ah thats cool. although you might get some false positives on the mmwave with speaker vibrations?
i took this pic before soldering the mic but here it is
cool
assuming your led section is correct you can add that back
ok
from the most part from what i see in your orginal code, you had the right idea on placing the light calls
can add bits back in slowly and make sure it doesnt break
you might want fast and slow pulse for different things though
effects:
- pulse:
name: "Slow Pulse"
transition_length: 250ms
update_interval: 250ms
min_brightness: 50%
max_brightness: 100%
- pulse:
name: "Fast Pulse"
transition_length: 100ms
update_interval: 100ms
min_brightness: 50%
max_brightness: 100%
there is an example HERE for the atom echo. which you can maybe take some inspiration from.
although its got some extra stuff which you wont need. but gives you an idea on how some devices work.
it uses some different components and only a single i2s bus so you cant directly copy much but gives you an example of some stuff
ok will take a look
once you get the hang of the basics with esphome it starts slotting together nicely
i have made lot of esphome devices like i made a custom smart switch it is fractions of the cost compared to other switchs
btw, how do I make the microphone listen from a distance and a little more accurately
i am not sure i understand what you mean by distance...
but you are only going to get a limited accuracy from a single INMP441
like from a far distance, it should hear my voice
any better mic for it?
with single mic you are always going to be limited. voice satelites normally overcome this with a multi mic array but then you probably need dedicate audio processing chip
you might be able to get some better results with other mics but realistically with a single mic and no dedicated audio processing you are going to be a bit limited
i did this because it is costly to import these or i would have bought it
i find my custom thing mostly works for me at the desk. but falls off fast. the voice pe's i have work more consistantly at a better range but i think they are only specced for 3.5m
i understand. realistically your going to be a bit limited. i like my setup for random testing but i probably wont use it for "production"
yes
i am going to make a v2 board of this with an actual PCB, 2 microphones, an xmos chip, all of it and mostly make it open source. I have multiple projects I want to do, but I am broke
also, am I able to play music on this device Just curious. With the help of a music assistant i tried but it just give me an error
[16:27:12][E][speaker_media_player:339]: The announcement pipeline's file reader encountered an error. [16:27:12][E][speaker_media_player.pipeline:112]: Media reader encountered an error: ESP_FAIL
i have seen simalar errors at points, i am not sure what causes it