Spaces:
Sleeping
Sleeping
File size: 2,067 Bytes
c41e64b 9bdf0c3 c41e64b 9bdf0c3 c41e64b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 |
from fastapi import FastAPI, File, UploadFile, HTTPException
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
import torch
import uvicorn
from fastapi.middleware.cors import CORSMiddleware
# Initialize FastAPI
app = FastAPI()
app.add_middleware(
CORSMiddleware,
allow_origins=['*'],
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
# Load the model and processor
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"
# Check if model exists locally, otherwise download it
if not os.path.exists(f"./{model_id}"):
model = AutoModelForSpeechSeq2Seq.from_pretrained(model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True)
processor = AutoProcessor.from_pretrained(model_id)
else:
model = AutoModelForSpeechSeq2Seq.from_pretrained(f"./{model_id}", torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True)
processor = AutoProcessor.from_pretrained(f"./{model_id}")
model.to(device)
pipe = pipeline(
"automatic-speech-recognition",
model=model,
tokenizer=processor.tokenizer,
feature_extractor=processor.feature_extractor,
torch_dtype=torch_dtype,
device=device,
)
# API endpoint to upload audio and get the transcribed text
@app.post("/transcribe")
async def transcribe_audio(file: UploadFile = File(...)):
try:
# Read the audio file bytes directly from the uploaded file
audio_bytes = await file.read()
# Pass the raw audio bytes to the pipeline
result = pipe(audio_bytes)
# Return the transcribed text
return {"text": result["text"]}
except Exception as e:
raise HTTPException(status_code=500, detail=f"Error occurred: {str(e)}")
@app.get("/")
async def root():
return {"message": "Welcome to the speech-to-text API!"}
# Running FastAPI with Uvicorn
if __name__ == "__main__":
uvicorn.run(app, host="0.0.0.0", port=8000)
|