app.py

import torch
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
# from datasets import load_dataset


device = "cuda:0" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32

model_id = "distil-whisper/distil-large-v3"

model = AutoModelForSpeechSeq2Seq.from_pretrained(
    model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
)
model.to(device)

processor = AutoProcessor.from_pretrained(model_id)

pipe = pipeline(
    "automatic-speech-recognition",
    model=model,
    tokenizer=processor.tokenizer,
    feature_extractor=processor.feature_extractor,
    max_new_tokens=128,
    torch_dtype=torch_dtype,
    device=device,
)

# Function to process audio input and transcribe it
@spaces.GPU
def transcribe(audio):
    # Load and preprocess the audio
    result = pipe(sample)
    return result["text"]

# Gradio interface
interface = gr.Interface(
    fn=transcribe, 
    inputs=gr.Audio(sources="microphone", type="filepath"),  
    outputs="text",
    title="Whisper Voice Transcription with Hugging Face"
)

# Launch the app
interface.launch()