update for hf

README.md

@@ -10,6 +10,8 @@ sdk_version: 5.9.0
 python_version: 3.10
 preload_from_hub:
   - FunAudioLLM/CosyVoice2-0.5B
+  - FunAudioLLM/SenseVoiceSmall
+  - FunAudioLLM/CosyVoice-ttsfrd
 ---
 
 [![SVG Banners](https://svg-banners.vercel.app/api?type=origin&text1=CosyVoice🤠&text2=Text-to-Speech%20💖%20Large%20Language%20Model&width=800&height=210)](https://github.com/Akshay090/svg-banners)

app.py

@@ -45,6 +45,7 @@ import numpy as np
 import torchaudio
 import random
 import librosa
+import spaces
 from funasr import AutoModel
 from funasr.utils.postprocess_utils import rich_transcription_postprocess
 ROOT_DIR = os.path.dirname(os.path.abspath(__file__))
@@ -97,6 +98,7 @@ def prompt_wav_recognition(prompt_wav):
     text = res[0]["text"].split('|>')[-1]
     return text
 
+@spaces.GPU
 def generate_audio(tts_text, mode_checkbox_group, prompt_text, prompt_wav_upload, prompt_wav_record, instruct_text,
                    seed, stream):
     sft_dropdown, speed = '', 1.0