Spaces:
Sleeping
Sleeping
Update Logic/SpeachToText.py
Browse files- Logic/SpeachToText.py +38 -17
Logic/SpeachToText.py
CHANGED
|
@@ -1,19 +1,40 @@
|
|
| 1 |
-
from fastapi import
|
| 2 |
-
from
|
|
|
|
| 3 |
import speech_recognition as sr
|
| 4 |
-
from
|
| 5 |
-
from
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
-
|
| 19 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from fastapi import FastAPI, File, UploadFile, HTTPException
|
| 2 |
+
from fastapi.responses import JSONResponse
|
| 3 |
+
from pydantic import BaseModel
|
| 4 |
import speech_recognition as sr
|
| 5 |
+
from io import BytesIO
|
| 6 |
+
from pydub import AudioSegment
|
| 7 |
+
import os
|
| 8 |
+
|
| 9 |
+
app = FastAPI()
|
| 10 |
+
|
| 11 |
+
class TranscriptionResponse(BaseModel):
|
| 12 |
+
text: str
|
| 13 |
+
|
| 14 |
+
@app.post("/transcribe", response_model=TranscriptionResponse)
|
| 15 |
+
async def transcribe_audio(file: UploadFile = File(...)):
|
| 16 |
+
if file.content_type not in ["audio/wav", "audio/mpeg", "audio/mp3", "audio/x-wav", "audio/flac"]:
|
| 17 |
+
raise HTTPException(status_code=400, detail="Unsupported file type")
|
| 18 |
+
|
| 19 |
+
try:
|
| 20 |
+
# Read the file into bytes
|
| 21 |
+
audio_data = await file.read()
|
| 22 |
+
audio_file = BytesIO(audio_data)
|
| 23 |
+
|
| 24 |
+
# Use pydub to handle different audio formats
|
| 25 |
+
audio = AudioSegment.from_file(audio_file, format=file.filename.split('.')[-1])
|
| 26 |
+
wav_audio = BytesIO()
|
| 27 |
+
audio.export(wav_audio, format="wav")
|
| 28 |
+
wav_audio.seek(0)
|
| 29 |
+
|
| 30 |
+
# Use speech_recognition to process the audio
|
| 31 |
+
recognizer = sr.Recognizer()
|
| 32 |
+
with sr.AudioFile(wav_audio) as source:
|
| 33 |
+
audio = recognizer.record(source)
|
| 34 |
|
| 35 |
+
# Recognize speech using Google Web Speech API
|
| 36 |
+
text = recognizer.recognize_google(audio, language="en-US")
|
| 37 |
+
return TranscriptionResponse(text=text)
|
| 38 |
+
|
| 39 |
+
except Exception as e:
|
| 40 |
+
raise HTTPException(status_code=500, detail=str(e))
|