src/transcription.py

"""Audio transcription using OpenAI Whisper API."""

import os
from openai import OpenAI


def get_openai_client() -> OpenAI:
    """Get OpenAI client with API key from environment."""
    api_key = os.getenv("OPENAI_API_KEY")
    if not api_key:
        raise ValueError("OPENAI_API_KEY environment variable not set")
    return OpenAI(api_key=api_key)


def transcribe_audio(audio_path: str) -> str:
    """
    Transcribe audio file using OpenAI Whisper API.

    Args:
        audio_path: Path to audio file (mp3, wav, webm, etc.)

    Returns:
        Transcribed text
    """
    client = get_openai_client()

    with open(audio_path, "rb") as audio_file:
        transcription = client.audio.transcriptions.create(
            model="whisper-1",
            file=audio_file,
            response_format="text"
        )

    return transcription


def transcribe_audio_chunk(audio_data: bytes, filename: str = "audio.webm") -> str:
    """
    Transcribe audio data directly from bytes.

    Args:
        audio_data: Raw audio bytes
        filename: Filename hint for format detection

    Returns:
        Transcribed text
    """
    client = get_openai_client()

    transcription = client.audio.transcriptions.create(
        model="whisper-1",
        file=(filename, audio_data),
        response_format="text"
    )

    return transcription