post-processing tech on the generated waves about lpcnet

[candlewill]

For example, from Baidu TTS online demo, it can be found from the spectrogram the audio is enhanced apparently by a post-processing tech to reduce the noise.

Baidu-TTS-sample.zip

[hyzhan]

@candlewill What about dataset size of "e2e_lpcnet_samples_share.zip". It sounds well. My e2e_demo have some noise, and loss about 3.35 for default parameters.

[OswaldoBornemann]

@candlewill Do you use tts with LPCNet ?

[candlewill]

yes

[OswaldoBornemann]

@candlewill Does LPCNet perform better than Griffin Lim on audio sound quality and audio generation speed ?

[candlewill]

@tsungruihon I think the LPCNet (train on your own dataset) is better than Griffin Lim both on quality and speed, especially when some post-processing algorithm could be applied.

[bearlu007]

@candlewill I think the synthesized sample you shared is very good actually. Probably just a little bit over fitting for LPCNet. By the way, I have the problem of training a stable LPCNet for TTS, I found LPCNet hard to converge, any suggestions ? Thanks

[OswaldoBornemann]

@candlewill have you ever tried LPCNet with 22050kHz ? thanks

[superhg2012]

@candlewill
have you tried any post processing techs to import sound quality?

[WhiteFu]

有没有什么好的后处理算法技术推荐的呢 @candlewill ，我用LPCnet的训练的时候总是有一些背景音。

[superhg2012]

有没有什么好的后处理算法技术推荐的呢 @candlewill ，我用LPCnet的训练的时候总是有一些背景音。

any progress?

[chengshaodi]

@superhg2012 您好，麻烦问一下，合成时会有一些噪音，您有解决办法吗？