no message

README.md

@@ -1,11 +1,9 @@
 ---
-license: apache-2.0
 title: CosyVoice
 sdk: gradio
 emoji: 🏃
 colorFrom: yellow
 colorTo: green
-python_version: 3.8.20
-sdk_version: 4.44.0
-app_file: deploy.py
+python_version: 3.8.9
+app_file: webui.py
 ---

deploy.sh

@@ -3,23 +3,23 @@ wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
 sh Miniconda3-latest-Linux-x86_64.sh -b
 source ~/miniconda3/bin/activate
 
-conda create -n cosyvoice python=3.8
-conda activate cosyvoice
-conda install -y -c conda-forge pynini==2.1.5
+#conda create -n cosyvoice python=3.8
+#conda activate cosyvoice
+#conda install -y -c conda-forge pynini==2.1.5
 # pynini is required by WeTextProcessing, use conda to install it as it can be executed on all platform.
 pip install -r requirements_.txt
 
 # If you encounter sox compatibility issues
 # ubuntu
 apt-get -y update
-apt-get install sox libsox-dev
+apt-get -y install sox libsox-dev
 
 mkdir -p pretrained_models
-git clone https://www.modelscope.cn/iic/CosyVoice-300M.git pretrained_models/CosyVoice-300M
+#git clone https://www.modelscope.cn/iic/CosyVoice-300M.git pretrained_models/CosyVoice-300M
 #git clone https://www.modelscope.cn/iic/CosyVoice-300M-25Hz.git pretrained_models/CosyVoice-300M-25Hz
 #git clone https://www.modelscope.cn/iic/CosyVoice-300M-SFT.git pretrained_models/CosyVoice-300M-SFT
 #git clone https://www.modelscope.cn/iic/CosyVoice-300M-Instruct.git pretrained_models/CosyVoice-300M-Instruct
-git clone https://www.modelscope.cn/iic/CosyVoice-ttsfrd.git pretrained_models/CosyVoice-ttsfrd
+#git clone https://www.modelscope.cn/iic/CosyVoice-ttsfrd.git pretrained_models/CosyVoice-ttsfrd
 #huggingface-cli download model-scope/CosyVoice-300M --local-dir pretrained_models/CosyVoice-300M --token=$hf_token
 #huggingface-cli download model-scope/CosyVoice-300M-SFT --local-dir pretrained_models/CosyVoice-300M-SFT --token=$hf_token
 #huggingface-cli download FunAudioLLM/CosyVoice-ttsfrd --local-dir pretrained_models/CosyVoice-ttsfrd --token=$hf_token
@@ -30,4 +30,6 @@ cd pretrained_models/CosyVoice-ttsfrd/
 unzip resource.zip -d .
 pip install ttsfrd-0.3.6-cp38-cp38-linux_x86_64.whl
 
-export PYTHONPATH=third_party/Matcha-TTS
+export PYTHONPATH=third_party/Matcha-TTS
+
+python3 webui.py

webui.py

@@ -132,7 +132,7 @@ def generate_audio(tts_text, mode_checkbox_group, sft_dropdown, prompt_text, pro
             yield (target_sr, i['tts_speech'].numpy().flatten())
 
 
-def main(args,sft_spk):
+def main():
     with gr.Blocks() as demo:
         gr.Markdown("### 代码库 [CosyVoice](https://github.com/FunAudioLLM/CosyVoice) \
                     预训练模型 [CosyVoice-300M](https://www.modelscope.cn/models/iic/CosyVoice-300M) \
@@ -171,18 +171,24 @@ def main(args,sft_spk):
     demo.launch(server_name='0.0.0.0', server_port=args.port)
 
 
-if __name__ == '__main__':
-    parser = argparse.ArgumentParser()
-    parser.add_argument('--port',
-                        type=int,
-                        default=8000)
-    parser.add_argument('--model_dir',
-                        type=str,
-                        default='pretrained_models/CosyVoice-300M',
-                        help='local path or modelscope repo id')
-    args = parser.parse_args()
-    cosyvoice = CosyVoice(args.model_dir)
-    sft_spk = cosyvoice.list_avaliable_spks()
-    prompt_sr, target_sr = 16000, 22050
-    default_data = np.zeros(target_sr)
-    main(args)
+
+# SDK模型下载
+from modelscope import snapshot_download
+snapshot_download('iic/CosyVoice-300M', local_dir='pretrained_models/CosyVoice-300M')
+
+os.system('apt-get -y update && apt-get -y install sox libsox-dev')
+
+parser = argparse.ArgumentParser()
+parser.add_argument('--port',
+                    type=int,
+                    default=8000)
+parser.add_argument('--model_dir',
+                    type=str,
+                    default='pretrained_models/CosyVoice-300M',
+                    help='local path or modelscope repo id')
+args = parser.parse_args()
+cosyvoice = CosyVoice(args.model_dir)
+sft_spk = cosyvoice.list_avaliable_spks()
+prompt_sr, target_sr = 16000, 22050
+default_data = np.zeros(target_sr)
+main()