合成音频像是电子音

使用my_run8_25k.pt模型生成音频，但合成后的音频听起来就像是电子音
看到运行时有报错说：UserWarning: nn.functional.tanh is deprecated. Use torch.tanh instead.
不知道会不会和这个有关系
已尝试把代码中所有提到nn.functional.tanh替换为torcht.tanh代替，仍有以上报错

下附完整运行日志：
```
D:\MockingBird-main>python web.py
Loaded synthesizer models: 3
Loaded encoder "pretrained.pt" trained to step 1564501
Building hifigan
Loading 'vocoder\saved_models\pretrained\g_hifigan.pt'
Complete.
Removing weight norm...
Web server:http://localhost:8080
127.0.0.1 - - [2021-10-31 11:52:20] "GET / HTTP/1.1" 200 15925 0.012005
127.0.0.1 - - [2021-10-31 11:52:20] "GET /api/synthesizers/ HTTP/1.1" 200 365 0.000996
Synthesizer using device: cpu
using synthesizer model: synthesizer\saved_models\mandarin\my_run8_25k.pt
Trainable Parameters: 32.738M
Loaded synthesizer "my_run8_25k.pt" trained to step 25000
+----------+---+
| Tacotron | r |
+----------+---+
| 25k | 2 |
+----------+---+

Read ['测试']
Synthesizing ['ce4 shi4']

| Generating 1/1
C:\ProgramData\Anaconda3\lib\site-packages\torch\nn\functional.py:1795: UserWarning: nn.functional.tanh is deprecated. Use torch.tanh instead.
warnings.warn("nn.functional.tanh is deprecated. Use torch.tanh instead.")

Done.

127.0.0.1 - - [2021-10-31 11:52:26] "POST /api/synthesize HTTP/1.1" 200 45150 3.512814
```
[合成音频](https://upload.07future.com/userUpLoad/5f6430f83aa26b564c01459f/xow31pUVn.wav)

The report runs web.py and loads synthesizer\saved_models\mandarin\my_run8_25k.pt, with a warning from torch.nn.functional.py. Start by tracing the synthesis path from web.py and checking whether the deprecated tanh call is related to the electronic-sounding output. Done means identifying the cause and verifying a correction or model incompatibility.

Written by the indexing model from the issue text.