Update app.py

app.py

@@ -1,5 +1,5 @@
 import os
-from fastapi import FastAPI, HTTPException, File, UploadFile, Depends, Security
+from fastapi import FastAPI, HTTPException, File, UploadFile, Depends, Security, Form
 from fastapi.security.api_key import APIKeyHeader, APIKey
 from fastapi.responses import JSONResponse
 from pydantic import BaseModel
@@ -70,9 +70,6 @@ class TTSRequest(BaseModel):
 class LanguageRequest(BaseModel):
     language: Optional[str] = None
 
-class TranscribeFileRequest(BaseModel):
-    language: Optional[str] = None
-
 async def get_api_key(api_key_header: str = Security(api_key_header)):
     if api_key_header == API_KEY:
         return api_key_header
@@ -165,7 +162,11 @@ async def transcribe_audio(request: AudioRequest, api_key: APIKey = Depends(get_
         )
 
 @app.post("/transcribe_file")
-async def transcribe_audio_file(file: UploadFile = File(...), request: TranscribeFileRequest = Depends(), api_key: APIKey = Depends(get_api_key)):
+async def transcribe_audio_file(
+    file: UploadFile = File(...),
+    language: Optional[str] = Form(None),
+    api_key: APIKey = Depends(get_api_key)
+):
     start_time = time.time()
     try:
         contents = await file.read()
@@ -178,12 +179,12 @@ async def transcribe_audio_file(file: UploadFile = File(...), request: Transcrib
         if sample_rate != ASR_SAMPLING_RATE:
             audio_array = librosa.resample(audio_array, orig_sr=sample_rate, target_sr=ASR_SAMPLING_RATE)
 
-        if request.language is None:
+        if language is None:
             # If no language is provided, use language identification
             identified_language = identify(audio_array)
             result = transcribe(audio_array, identified_language)
         else:
-            result = transcribe(audio_array, request.language)
+            result = transcribe(audio_array, language)
 
         processing_time = time.time() - start_time
         return JSONResponse(content={"transcription": result, "processing_time_seconds": processing_time})
@@ -321,7 +322,10 @@ async def identify_language(request: AudioRequest, api_key: APIKey = Depends(get
         )
 
 @app.post("/identify_file")
-async def identify_language_file(file: UploadFile = File(...), api_key: APIKey = Depends(get_api_key)):
+async def identify_language_file(
+    file: UploadFile = File(...),
+    api_key: APIKey = Depends(get_api_key)
+):
     start_time = time.time()
     try:
         contents = await file.read()
@@ -339,7 +343,7 @@ async def identify_language_file(file: UploadFile = File(...), api_key: APIKey =
         return JSONResponse(
             status_code=500,
             content={"message": "An error occurred during language identification", "details": error_details, "processing_time_seconds": processing_time}
-        )
+        # ... (previous code remains the same)
 
 @app.post("/asr_languages")
 async def get_asr_languages(request: LanguageRequest, api_key: APIKey = Depends(get_api_key)):
@@ -388,3 +392,26 @@ async def get_tts_languages(request: LanguageRequest, api_key: APIKey = Depends(
             status_code=500,
             content={"message": "An error occurred while fetching TTS languages", "details": error_details, "processing_time_seconds": processing_time}
         )
+
+# If you want to add a health check endpoint
+@app.get("/health")
+async def health_check():
+    return {"status": "ok"}
+
+# You might also want to add a root endpoint that provides basic API information
+@app.get("/")
+async def root():
+    return {
+        "message": "Welcome to the MMS Speech Technology API",
+        "version": "1.0",
+        "endpoints": [
+            "/transcribe",
+            "/transcribe_file",
+            "/synthesize",
+            "/identify",
+            "/identify_file",
+            "/asr_languages",
+            "/tts_languages",
+            "/health"
+        ]
+    }