-
Hello I am using ESPNet2 and trying to feed the mel spectrogram I get from FS2 into different vocoders (not PWG). But the output of the spectrogram is not going into the other vocoders properly even i…
gafsd updated
2 years ago
-
Hello !
I just rewrote the WaveGlow vocoder in tensorflow trying to improve its performance but it fails (1.7 sec generated / sec with the pytorch model drops down to less than 1.2 generated / sec in…
ghost updated
2 years ago
-
Hi @kan-bayashi
A general question regarding the LibriTTS vocoders -- I am using them (melgan/hifigan/parallel_wavegan) to decode the spectrograms generated from LibriTTS Tacotron2/Transformer-TTS…
-
Hi, I am getting the following error
`from espnet_onnx.espnet_onnx.export import TTSModelExport
`
-
What are good ways of assessing quality achieved by the LPCNet Vocoder? I found that PESQ results are way worse than the quality I experienced myself, with ratings constantly in the range between 1 an…
-
```
| Hparams chains: ['training/config.yaml']
| Hparams:
;33;mK_step: 1000, ;33;maccumulate_grad_batches: 1, ;33;maudio_num_mel_bins: 80, ;33;maudio_sample_rate: 24000, ;33;mbinarization_args: {…
-
Related to **fastPitch/Pytorch**
**Describe the bug**
As the pitch is normalization by mean-variance normalization, the pitch with non-zero value is converted to zero-center. The pitch of unvoice…
-
Hi
Thanks for this great project.
I trained synthesizer using my data and used pretrained vocoder and the result was good. Now I want to use other vocoders. I cloned HiFi-GAN project and tried to ru…
-
环境
操作系统：window10
Anaconda版本：4.3
Python版本：3.7
运行：python demo_toolbox.py -d .\samples （可以正常运行）
运行：python web.py报错，错误信息如下：
(pythone37) E:\CACHE_DATA\CHROME\MockingBird-main-new\MockingBird-main>p…
-
微博内容精选