app.py

import spaces
import gradio as gr
import os
import httpx

@spaces.GPU(duration=0)
def transcribe(audio_path):
    if audio_path is None:
        return "Please upload an audio file."
    
    api_key = os.environ.get("ELEVENLABS_API_KEY")
    if not api_key:
        return "ERROR: ELEVENLABS_API_KEY not found!"
    
    try:
        with open(audio_path, "rb") as audio_file:
            response = httpx.post(
                "https://api.elevenlabs.io/v1/speech-to-text",
                headers={"xi-api-key": api_key},
                files={"file": audio_file},
                data={"model_id": "scribe_v1"},
                timeout=120
            )
        
        if response.status_code == 200:
            result = response.json()
            return result.get("text", "No text found")
        else:
            return f"ERROR {response.status_code}: {response.text}"
            
    except Exception as e:
        return f"ERROR: {str(e)}"

demo = gr.Interface(
    fn=transcribe,
    inputs=gr.Audio(type="filepath"),
    outputs=gr.Textbox(lines=20),
    title="Audio Transcription - ElevenLabs Scribe",
    description="Upload audio for accurate transcription."
)

demo.launch()