How to generate speech from features with WORLD vocoder

Question

How to generate speech from features with WORLD vocoder

ghostcow opened this issue 6 years ago · comments

Hi,

I'm trying to debug my system that uses you WaveNet vocoder. Is there any way to create WAV from the features your code generates?

Thanks

Tomoki Hayashi · Answer 1 · Sat Apr 21 2018 08:00:49 GMT+0800 (China Standard Time)

What does it mean?
You can run one of the recipes, then you can get generated wav file from feature vectors.

Lior Uzan · Answer 2 · Sat Apr 21 2018 08:05:08 GMT+0800 (China Standard Time)

I meant, by using the WORLD vocoder instead of the WaveNet.
My question essentially is, why do you change the F0 when creating the features? Is that really necessary?

Tomoki Hayashi · Answer 3 · Sat Apr 21 2018 08:30:52 GMT+0800 (China Standard Time)

I have tried both (f0 and continuous f0 + uv information), but I did not compare them strictly.
The reason why I use continuous f0 is based on our experiences.
Anyway, both work, therefore, maybe you can use f0 directly, instead of continuous f0 + uv.

And our feature extraction step is based on the sprocket, which use world in internal processing.
By using sprocket.speech.Synthesizer, you can generate voice with world vocoder.

Lior Uzan · Answer 4 · Sat Apr 21 2018 19:12:05 GMT+0800 (China Standard Time)

Thanks! I'll try your suggestions!