Ryuichi Yamamoto

I have created a jupyter notebook to demonstrate how we can use pre-trained models to generate singing voice samples. ### Neural network-based singing voice synthesis demo using kiritan_singing database (Japanese)...

I pushed all the code for feature extraction, training, and inference as well. Models used in the above demo can be reproduced by running the following recipe: https://github.com/r9y9/nnsvs/tree/master/egs/kiritan_singing/00-svs-world

I made a new recipe for **nit-song070**, which is a singing voice dataset provided by the [HTS working group](http://hts.sp.nitech.ac.jp/). The dataset contains 31 songs recorded by a female Japanese singer....

I have added another recipe for [jsut-song](https://sites.google.com/site/shinnosuketakamichi/publication/jsut-song) dataset. - Sample: https://soundcloud.com/r9y9/20200525-haru-ga-kita-5-jsut-song?in=r9y9/sets/dnn-based-singing-voice - Recipe: https://github.com/r9y9/nnsvs/tree/master/egs/jsut-song/00-svs-world

Good news: the author of NSF published a pytorch implementation of NSF: https://github.com/nii-yamagishilab/project-NN-Pytorch-scripts It should be easy to integrate it with our codebase.

`out_acoustic` directory contains 1) acoustic features (`*-feats.npy`) and 2) waveform (`*-wave.npy`), which can be used for training neural vocoders. ``` ls -l dump/kiritan/norm/train_no_dev/out_acoustic/ | head total 1254736 -rw-rw-r-- 1 ryuichi...

I would like to note that I plan to revisit the neural vocoder support after huge refactoring #81. I will be busy at work so It would take some time...

Let me rename the issue title.

Code is ready at the dev branch: https://github.com/r9y9/nnsvs/tree/dev Here's a pre-trained model using the [mixed/voc-multidb-latest](https://github.com/r9y9/nnsvs/tree/dev/recipes/mixed/voc-multidb-latest) recipe: https://www.dropbox.com/sh/52u8u5z41e4sdnz/AABTeOwtgyRiPqrAa5sjXFSHa?dl=0 I trained the model by: ``` ./run.sh --stage 4 --stop-stage 4 --vocoder-model hn-sinc-nsf_sr48k_pwgD...

TODO: GTA training support like: https://github.com/r9y9/ttslearn/blob/4939934554daadd0bd85c995e70d88f58206056e/extra_recipes/jsut/tacotron2_pwg/run.sh#L142-L167