app.py

import os
import logging
import gradio as gr
from huggingface_hub import InferenceClient

# Configure logging
logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s | %(levelname)s | %(message)s",
    datefmt="%Y-%m-%d %H:%M:%S",
)
logger = logging.getLogger(__name__)

# Environment variables for configuration
HF_TOKEN = os.environ.get("HF_TOKEN", "")
MODEL_ID = os.environ.get("MODEL_ID", "openai/whisper-large-v3-turbo")

logger.info(f"HF_TOKEN configured: {bool(HF_TOKEN)}")
logger.info(f"MODEL_ID: {MODEL_ID}")

client = InferenceClient(token=HF_TOKEN) if HF_TOKEN else InferenceClient()
logger.info("InferenceClient initialized")



def transcribe(audio) -> str:
    """Transcribe audio file or recording."""
    logger.info(f"transcribe() called | audio={audio}")
    
    if audio is None:
        logger.warning("No audio provided")
        return "🎤 Record or upload audio first!"
    
    try:
        logger.info(f"Calling automatic_speech_recognition | model={MODEL_ID}")
        result = client.automatic_speech_recognition(audio, model=MODEL_ID)
        logger.info(f"Transcription: {len(result.text)} chars")
        return result.text
    except Exception as e:
        logger.error(f"API error: {e}")
        return f"❌ Error: {e}"


logger.info("Building Gradio interface...")

with gr.Blocks(title="Whisper Transcriber") as demo:
    gr.Markdown("# 🎙️ Whisper Transcriber\nRecord your voice or upload an audio file to get a transcription.")

    with gr.Row(equal_height=True):
        with gr.Column():
            mic = gr.Audio(sources=["microphone"], type="filepath", label="🎤 Record")
        with gr.Column():
            upload = gr.Audio(sources=["upload"], type="filepath", label="📁 Upload")

    output = gr.Textbox(label="Transcription", lines=4, interactive=False)

    with gr.Row():
        btn_mic = gr.Button("Transcribe Recording", variant="primary")
        btn_upload = gr.Button("Transcribe Upload", variant="secondary")

    btn_mic.click(transcribe, inputs=mic, outputs=output)
    btn_upload.click(transcribe, inputs=upload, outputs=output)

demo.queue()
logger.info("Starting Gradio server...")
demo.launch()