tools/transcribe_audio.py

import os
from openai import OpenAI
from smolagents import tool


@tool
def transcribe_mp3_with_whisper(task_id:str)->str: # 10 minutes in milliseconds
    """
    Transcribes an MP3 audio file using OpenAI's Whisper API.
    Handles larger files by chunking them.

    Args:
        task_id (str): The task_id of the audio file to be transcribed.
      
    """
    client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY"))

    if not client.api_key:
        print("Error: OpenAI API key not found. Please set the OPENAI_API_KEY environment variable.")
        return

    file_path = os.path.join(task_id + ".mp3")
    
    full_transcript = ""

    with open(file_path, "rb") as audio_file:
        transcript = client.audio.transcriptions.create(
            model="whisper-1",
            file=audio_file
        )
        full_transcript = transcript.text

    return full_transcript.strip()

if __name__ == "__main__":
    # Replace with the path to your downloaded MP3 file
    input_mp3_file = os.path.join("1f975693-876d-457b-a649-393859e79bf3")


    print(f"Transcribing '{input_mp3_file}'...")

    text = transcribe_mp3_with_whisper(input_mp3_file)
    print(f"Transcription completed. Output saved to '{text}'.")