#Python Integration with an OpenAI Custom GPT.

1 messages · Page 1 of 1 (latest)

grave kayak
#

Title kinda says it.

Has anyone connected a Custom GPT to ElevenLabs to generate audio with a bit more control? E.g. speech speed?

wheat bramble
#

There is no way to control the speed of the audio generated with the ElevenLabs API at the moment and even if there was I don't think a GPT can send external API calls and receive the response.. EDIT: I mean to API's not already in their "Integration list"

grave kayak
#

So what's this ') VoiceSettings(speaking_rate=0.8)

and why does my OpenAI assist write python code to make api calls?

wheat bramble
#

So what's this ') VoiceSettings(speaking_rate=0.8)
It is not a valid paramater for the ElevenLabs API so I don't know. Where did you get the code from?

and why does my OpenAI assist write python code to make api calls?
It runs that python code "locally".

grave kayak
#

Got it from an ElevenLabs docs

wheat bramble
#

Can you screenshot and/or link the documentation you are referring to?
This is the standard TTS call in python. There is no "speed" parameter unfortunately.


url = "https://api.elevenlabs.io/v1/text-to-speech/{voice_id}"

payload = {
    "model_id": "<string>",
    "pronunciation_dictionary_locators": [
        {
            "pronunciation_dictionary_id": "<string>",
            "version_id": "<string>"
        }
    ],
    "text": "<string>",
    "voice_settings": {
        "similarity_boost": 123,
        "stability": 123,
        "style": 123,
        "use_speaker_boost": True
    }
}
headers = {"Content-Type": "application/json"}

response = requests.request("POST", url, json=payload, headers=headers)

print(response.text)```
unkempt pollen
#

that's seriously still there?

#

yeah that part of the documentation is just wrong

wheat bramble
#

that's seriously still there?
Yeah I know, I was just using it as an example that there is no Speed parameter ..

unkempt pollen
#

it's a known error on one of the doc pages