newCommit

main.py

@@ -2,10 +2,7 @@ from fastapi import FastAPI, File, UploadFile, HTTPException
 from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
 import torch
 import uvicorn
-import soundfile as sf
 from fastapi.middleware.cors import CORSMiddleware
-import os
-import tempfile
 
 # Initialize FastAPI
 app = FastAPI()
@@ -46,31 +43,16 @@ pipe = pipeline(
 @app.post("/transcribe")
 async def transcribe_audio(file: UploadFile = File(...)):
     try:
-        # Create a temporary file to save the uploaded content
-        with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as temp_file:
-            temp_file.write(await file.read())
-            temp_path = temp_file.name
+        # Read the audio file bytes directly from the uploaded file
+        audio_bytes = await file.read()
 
-        # Load the audio file using soundfile
-        audio, sr = sf.read(temp_path)
-
-        # Ensure the sample rate is 16000 Hz
-        if sr != 16000:
-            raise HTTPException(status_code=400, detail="Sample rate must be 16000 Hz.")
-
-        # Pass the processed audio to the pipeline
-        result = pipe(temp_path)
-
-        # Remove the temp file after processing
-        os.remove(temp_path)
+        # Pass the raw audio bytes to the pipeline
+        result = pipe(audio_bytes)
 
         # Return the transcribed text
         return {"text": result["text"]}
     
     except Exception as e:
-        # Clean up temp file in case of error
-        if 'temp_path' in locals() and os.path.exists(temp_path):
-            os.remove(temp_path)
         raise HTTPException(status_code=500, detail=f"Error occurred: {str(e)}")
 
 @app.get("/")