Update main.py

main.py

@@ -1,9 +1,8 @@
 from fastapi import FastAPI, Form
-from fastapi.responses import FileResponse
+from fastapi.responses import StreamingResponse
+import io
 import wave
 from piper.voice import PiperVoice
-import os
-import tempfile
 
 app = FastAPI()
 
@@ -20,20 +19,21 @@ def synthesize_text(text: str = Form(...)):
         text (str): The text to synthesize.
 
     Returns:
-        FileResponse: The WAV file containing the synthesized speech.
+        StreamingResponse: The audio data as a stream.
     """
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as temp_file:
-        output_file = temp_file.name
+    audio_buffer = io.BytesIO()
 
-    # Synthesize speech and save to a WAV file
-    with wave.open(output_file, 'wb') as wav_file:
+    # Synthesize speech and write to an in-memory WAV file
+    with wave.open(audio_buffer, 'wb') as wav_file:
+        wav_file.setnchannels(1)  # Set the number of audio channels
+        wav_file.setsampwidth(2)  # Set sample width to 2 bytes
+        wav_file.setframerate(16000)  # Set the sampling rate
         audio = voice.synthesize(text, wav_file)
 
-    # Ensure the file exists and return it as a response
-    if os.path.exists(output_file):
-        return FileResponse(output_file, media_type='audio/wav', filename='output.wav')
-    else:
-        return {"error": "Failed to generate audio."}
+    # Seek to the beginning of the buffer so it can be read from the start
+    audio_buffer.seek(0)
+
+    return StreamingResponse(audio_buffer, media_type="audio/wav")
 
 @app.get("/")
 def read_root():