| from transformers import pipeline
|
| from typing import Dict, Any
|
|
|
|
|
|
|
| transcriber = pipeline("automatic-speech-recognition", model="distil-whisper/distil-small.en")
|
|
|
| async def send_to_voice_AI(audio_data: bytes) -> Dict[str, Any]:
|
| """
|
| Transcribes audio data using the Hugging Face Whisper model.
|
|
|
| Args:
|
| audio_data (bytes): The raw audio data to be transcribed.
|
|
|
| Returns:
|
| Dict[str, Any]: A dictionary containing the transcription result or error message.
|
| """
|
| print("Transcribing audio")
|
| try:
|
|
|
| result = transcriber(audio_data)
|
| return {"text": result["text"]}
|
| except Exception as e:
|
| print(f"Error during transcription: {str(e)}")
|
| return {"text": "Sorry, an error occurred during transcription."} |
