main.py

from fastapi import FastAPI, UploadFile, File
import torch
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline

app = FastAPI()


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

model_id = "openai/whisper-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,
    chunk_length_s=30,
    batch_size=16,
    return_timestamps=True,
    torch_dtype=torch_dtype,
    device=device,
)

sample = dataset[0]["audio"]

# result = pipe(sample)
# print(result["text"])


@app.post("/speech_to_text")
async def speech_to_text(file : UploadFile = File(...))
    if file:
        contents = await file.read()
        with open(file.filename, "wb") as f:
            f.write(contents)

        converted_result = pipe(file.filename)
        return {
            "status": 200,
            "text": converted_result["text"]
        }
    else:
        return {
            "status": -1
        }