LiveKit Community

what other options do i have for this:_input_audio_transcription_=InputAudioTranscription(
        _model_="whisper-1",
        _language_="tr"
    ),

Thanks for reaching out, Rhythm. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.


You have several options for configuring `InputAudioTranscription` beyond just `model="whisper-1"` and `language="tr"`:
• *Supported models:* 
    ◦ `gpt-4o-mini-transcribe` (default for non-Azure)
    ◦ `gpt-4o-transcribe`
    ◦ `whisper-1` (default for Azure)
• *Supported languages:* Use ISO-639-1 codes. Examples include `en` (English), `tr` (Turkish), `de` (German), `fr` (French), `es` (Spanish), `zh` (Chinese), `ru` (Russian), and many more. Regional variants like `en-US`, `pt-BR`, `fr-CA` are also supported.
• *Additional parameters:* 
    ◦ `prompt`: Optional, only supported for `whisper-1`
    ◦ `detect_language`: Enables automatic language detection
    ◦ `noise_reduction_type`: Set to `near_field` or `far_field` for noise reduction
Example configuration:
```InputAudioTranscription(
    model="gpt-4o-mini-transcribe",
    language="en-US",
    noise_reduction_type="near_field"
)```
Use the `prompt` parameter only with `whisper-1` if you want to provide a text prompt for context.
Let me know if you need details on a specific model, language, or parameter.
 Sources: <https://github.com/livekit/agents-js/blob/a1e5c498be2363c0d26779971f04d29b81f6920a/plugins/openai/src/stt.ts#L1-L72|plugins/openai/src/stt.ts> | <https://github.com/livekit/agents/blob/fd648b6c9c2f577c39a395a36585f152edac542d/livekit-plugins/livekit-plugins-google/livekit/plugins/google/models.py#L1-L106|livekit-plugins/livekit-plugins-google/livekit/plugins/google/models.py> | <https://docs.livekit.io/agents/integrations/stt/openai|OpenAI STT integration guide | LiveKit Docs>