Update app.py

app.py

@@ -8,7 +8,7 @@ import av
 import gradio as gr
 import spaces
 import torch
-from fastrtc import AdditionalOutputs, ReplyOnPause, WebRTC, WebRTCData, get_hf_turn_credentials
+from fastrtc import AdditionalOutputs, ReplyOnPause, WebRTC, WebRTCData, get_hf_turn_credentials, audio_to_int16
 from gradio.processing_utils import save_audio_to_cache
 from gradio.utils import get_upload_folder
 from transformers import AutoModelForImageTextToText, AutoProcessor
@@ -205,7 +205,7 @@ def generate(data: WebRTCData, history: list[dict], system_prompt: str = "", max
     files = []
     audio_file = None
     if data.audio is not None and data.audio[1].size > 0:
-        audio_file = save_audio_to_cache(data.audio[1], data.audio[0], format="mp3", cache_dir=get_upload_folder())
+        audio_file = save_audio_to_cache(audio_to_int16(data.audio[1]), data.audio[0], format="mp3", cache_dir=get_upload_folder())
         files.append(audio_file)
     if image is not None:
         files.append(image)