WaveRNN | WaveRNN Vocoder + TTS | Speech library

by fatchord Python Version: Current License: MIT

X-Ray Key Features Code Snippets(3)Community Discussions(1)Vulnerabilities Install Support

kandi X-RAY | WaveRNN Summary

WaveRNN is a Python library typically used in Artificial Intelligence, Speech, Pytorch applications. WaveRNN has no vulnerabilities, it has build file available, it has a Permissive License and it has medium support. However WaveRNN has 4 bugs. You can download it from GitHub.

Pytorch implementation of Deepmind's WaveRNN model from Efficient Neural Audio Synthesis.

Support

Quality

Security

License

Reuse

Support

WaveRNN has a medium active ecosystem.

It has 1953 star(s) with 666 fork(s). There are 86 watchers for this library.

It had no major release in the last 6 months.

There are 96 open issues and 126 have been closed. On average issues are closed in 64 days. There are 8 open pull requests and 0 closed requests.

It has a neutral sentiment in the developer community.

The latest version of WaveRNN is current.

Quality

WaveRNN has 4 bugs (2 blocker, 0 critical, 1 major, 1 minor) and 92 code smells.

Security

WaveRNN has no vulnerabilities reported, and its dependent libraries have no vulnerabilities reported.

WaveRNN code analysis shows 0 unresolved vulnerabilities.

There are 0 security hotspots that need review.

License

WaveRNN is licensed under the MIT License. This license is Permissive.

Permissive licenses have the least restrictions, and you can use them in most projects.

Reuse

WaveRNN releases are not available. You will need to build from source code and install.

Build file is available. You can build the component from source.

Installation instructions are available. Examples and code snippets are not available.

WaveRNN saves you 1042 person hours of effort in developing the same functionality from scratch.

It has 2364 lines of code, 182 functions and 30 files.

It has high code complexity. Code complexity directly impacts maintainability of the code.

Top functions reviewed by kandi - BETA

kandi has reviewed WaveRNN and discovered the below as its top functions. This is intended to give you an instant insight into WaveRNN implemented functionality, and help decide if they suit your requirements.

Generate the wave function
Fold a tensor with overlap
Sample from discretized mixture
Decode the value of the mu - law function
Performs training loop
Save a checkpoint
Helper function for training data
Discretized mixture logistic loss
Compute the logarithm of x
Save an attention matrix
Cleans up whitespace
Transliterate transliteration
Processes a WAV file
Clean English text
Reconstruct a waveform from a given mel file
Reads text from CSV file
Loads the vocab dataset
Parse the pronunciation file
Forward computation
Restore a checkpoint
Performs forward computation
Create a simple table from a list of tuples
Load TTS dataset
Generate a welspectrogram from a file
Collate the vocab
Convert text to sequence

Get all kandi verified functions for this library.

WaveRNN Key Features

No Key Features are available at this moment for WaveRNN.

WaveRNN Examples and Code Snippets

Low-latency real-time multispeaker voice conversion (VC) with cyclic variational autoencoder (CycleVAE) and multiband WaveRNN using data-driven linear prediction (MWDLP),Installation

Python

Lines of Code : 3

License : Permissive (Apache-2.0)

Copy

$ cd tools
$ make
$ cd ..

Tacotron 2 - Persian,Demo

Python

Lines of Code : 1

License : No License

Copy

sh scripts/demo/generate.sh

Training,Citation,Neural vocoder (WaveRNN)

Python

Lines of Code : 1

License : Permissive (MIT)

Copy

CUDA_VISIBLE_DEVICES=0 python vocoder_train.py -g --syn_dir datasets/vctk/synthesizer vctk datasets/vctk

Community Discussions

Trending Discussions on WaveRNN

How to reproduce RNN results on several runs?

QUESTION

How to reproduce RNN results on several runs?

Asked 2019-May-20 at 03:44

I call same model on same input twice in a row and I don't get the same result, this model have nn.GRU layers so I suspect that it have some internal state that should be release before second run?

How to reset RNN hidden state to make it the same as if model was initially loaded?

UPDATE:

Some context:

I'm trying to run model from here:

https://github.com/erogol/WaveRNN/blob/master/models/wavernn.py#L93

I'm calling generate:

https://github.com/erogol/WaveRNN/blob/master/models/wavernn.py#L148

Here it's actually have some code using random generator in pytorch:

https://github.com/erogol/WaveRNN/blob/master/models/wavernn.py#L200

https://github.com/erogol/WaveRNN/blob/master/utils/distribution.py#L110

https://github.com/erogol/WaveRNN/blob/master/utils/distribution.py#L129

I have placed (I'm running code on CPU):

...

ANSWER

Answered 2019-May-18 at 00:09

I believe this may be highly related to Random Seeding. To ensure reproducible results (as stated by them) you have to seed torch as in this:

Source https://stackoverflow.com/questions/56190274

Community Discussions, Code Snippets contain sources that include Stack Exchange Network

Vulnerabilities

No vulnerabilities reported

Install WaveRNN

Then install the rest with pip:. pip install -r requirements.txt.
Python >= 3.6
Pytorch 1 with CUDA
If you want to use TTS functionality immediately you can simply use:. This will generate everything in the default sentences.txt file and output to a new 'quick_start' folder where you can playback the wav files and take a look at the attention plots.

Support

For any new features, suggestions and bugs create an issue on GitHub. If you have any questions check and ask questions on community page Stack Overflow .

Find more information at: