zillaru / emotivoice-tpu Goto Github PK

报错信息——requests.exceptions.ConnectionError: HTTPSConnectionPool(host='disk.sophgo.vip', port=443): Max retries exceeded with url: /sharing/webapi/entry.cgi?api=SYNO.Core.Sharing.Session&version=1&method=get&sharing_id=%22KymDuWLGw%22 (Caused by NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7f8a049400>: Failed to establish a new connection: [Errno 110] Connection timed out'))

I fail to download models by "python3 -m dfn --url https://disk.sophgo.vip/sharing/KymDuWLGw".
I try to download models by baidu network disk and google network disk，both files look the same.

The checkpoints directory is organized as:
.
├── base_speakers
│   ├── EN
│   │   ├── checkpoint.pth
│   │   ├── config.json
│   │   ├── decoder_1024_f16.bmodel
│   │   ├── decoder_1024_f32.bmodel
│   │   ├── en_default_se.pth
│   │   └── en_style_se.pth
│   └── ZH
│   ├── checkpoint.pth
│   ├── config.json
│   ├── decoder_1024_f16.bmodel
│   └── zh_default_se.pth
└── converter
├── checkpoint.pth
├── config.json
├── decoder_1024_f16.bmodel
└── decoder_2048_f16.bmodel

how to download tts models as follow:
model_file
└── tts
├── am_durationpred-1_512_384-1_512_1.bmodel
├── am_energypred-1_512_384-1_512_1.bmodel
├── am_pitchpred-1_512_384-1_512_1.bmodel
├── am_rest_weight.pth
├── bert_poolout1-768_1-512_1-512_1-512.bmodel
├── hifigan_1-80-1024_F16.bmodel
├── jit_am_encoder_1-512-384_1-1-512.bmodel
└── onnx_am_decoder_1-2048-384.bmodel

How to convert models to bmodel? Can you provide a reference example? Thanks.

miss output_dir in convert_only in demo_page.py line 219

save_path = f'{output_dir}/output.wav'

new version of gradio Audio component without info attribution,
if use gradio==4.19.2 the audio support multi sources ['upload', 'microphone'] which can be friendly to use user tone color

for the Error Related to Silero I reference this page https://github.com/myshell-ai/OpenVoice/blob/main/docs/QA.md#issues-with-installation

求助最后一步执行demopage.py的时候报错请问如何解决呀