Update app.py

app.py

@@ -1,11 +1,11 @@
-from fastapi import FastAPI, UploadFile, File, Form
+from fastapi import FastAPI, Form
 from fastapi.responses import JSONResponse, FileResponse
 import uvicorn
 from pydantic import BaseModel
 import numpy as np
 import io
 import soundfile as sf
-
+import base64
 from asr import transcribe, ASR_LANGUAGES
 from tts import synthesize, TTS_LANGUAGES
 from lid import identify
@@ -17,12 +17,16 @@ class TTSRequest(BaseModel):
     language: str
     speed: float
 
+class AudioRequest(BaseModel):
+    audio: str  # Base64 encoded audio data
+    language: str
+
 @app.post("/transcribe")
-async def transcribe_audio(audio: UploadFile = File(...), language: str = Form(...)):
-    contents = await audio.read()
-    audio_array, sample_rate = sf.read(io.BytesIO(contents))
+async def transcribe_audio(request: AudioRequest):
+    audio_bytes = base64.b64decode(request.audio)
+    audio_array, sample_rate = sf.read(io.BytesIO(audio_bytes))
     
-    result = transcribe(audio_array, language)
+    result = transcribe(audio_array, request.language)
     return JSONResponse(content={"transcription": result})
 
 @app.post("/synthesize")
@@ -41,9 +45,9 @@ async def synthesize_speech(request: TTSRequest):
     )
 
 @app.post("/identify")
-async def identify_language(audio: UploadFile = File(...)):
-    contents = await audio.read()
-    audio_array, sample_rate = sf.read(io.BytesIO(contents))
+async def identify_language(request: AudioRequest):
+    audio_bytes = base64.b64decode(request.audio)
+    audio_array, sample_rate = sf.read(io.BytesIO(audio_bytes))
     
     result = identify(audio_array)
     return JSONResponse(content={"language_identification": result})
@@ -54,4 +58,4 @@ async def get_asr_languages():
 
 @app.get("/tts_languages")
 async def get_tts_languages():
-    return JSONResponse(content=TTS_LANGUAGES)
+    return JSONResponse(content=TTS_LANGUAGES)