app.py

import gradio as gr
import torch
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
from datasets import load_dataset
import soundfile as sf

# Check if CUDA is available and set the device
device = "cuda:0" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32

# Load the model and processor
model_id = "openai/whisper-large-v3"
model = AutoModelForSpeechSeq2Seq.from_pretrained(
    model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=False, use_safetensors=True
).to(device)
processor = AutoProcessor.from_pretrained(model_id)

# Define the ASR pipeline
asr_pipeline = pipeline(
    "automatic-speech-recognition",
    model=model,
    tokenizer=processor.tokenizer,
    feature_extractor=processor.feature_extractor,
    max_new_tokens=128,
    chunk_length_s=30,
    batch_size=16,
    return_timestamps=True,
    torch_dtype=torch_dtype,
    device=device,
)

# Function to process audio in chunks and return the combined text
def process_audio(file_info):
    path = file_info["path"]
    audio_stream = sf.SoundFile(path, 'r')
    results = []
    while True:
        data = audio_stream.read(dtype='float32')
        if len(data) == 0:
            break
        result = asr_pipeline(data)
        results.append(result)
    audio_stream.close()
    combined_text = " ".join([r["text"] for r in results])
    return combined_text

# Create the Gradio interface
iface = gr.Interface(
    fn=process_audio,
    inputs=gr.Audio(label="Upload your audio file or use the microphone"),
    outputs="text",
    title="👋🏻Welcome To 🙋🏻‍♂️Patrick's Whisper🌬️",
    description=""""
You can use this Space to test out the current model [Whisper3Large](https://huggingface.co/openai/whisper-large-v3)
You can also use 🙋🏻‍♂️Patrick's Whisper🌬️ by cloning this space. 🧬🔬🔍 Simply click here: <a style="display:inline-block" href="https://huggingface.co/spaces/PatsWhisper3Large?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a></h3> 
Join us : 🌟TeamTonic🌟 is always making cool demos! Join our active builder's🛠️community on 👻Discord: [Discord](https://discord.gg/GWpVpekp) On 🤗Huggingface: [TeamTonic](https://huggingface.co/TeamTonic) & [MultiTransformer](https://huggingface.co/MultiTransformer) On 🌐Github: [Polytonic](https://github.com/tonic-ai) & contribute to 🌟 [PolyGPT](https://github.com/tonic-ai/polygpt-alpha)
"""
)

# Launch the application
iface.launch()