A TensorFlow implementation of DeepMind's WaveNet paper

Overview

A TensorFlow implementation of DeepMind's WaveNet paper

Build Status

This is a TensorFlow implementation of the WaveNet generative neural network architecture for audio generation.

The WaveNet neural network architecture directly generates a raw audio waveform, showing excellent results in text-to-speech and general audio generation (see the DeepMind blog post and paper for details).

The network models the conditional probability to generate the next sample in the audio waveform, given all previous samples and possibly additional parameters.

After an audio preprocessing step, the input waveform is quantized to a fixed integer range. The integer amplitudes are then one-hot encoded to produce a tensor of shape (num_samples, num_channels).

A convolutional layer that only accesses the current and previous inputs then reduces the channel dimension.

The core of the network is constructed as a stack of causal dilated layers, each of which is a dilated convolution (convolution with holes), which only accesses the current and past audio samples.

The outputs of all layers are combined and extended back to the original number of channels by a series of dense postprocessing layers, followed by a softmax function to transform the outputs into a categorical distribution.

The loss function is the cross-entropy between the output for each timestep and the input at the next timestep.

In this repository, the network implementation can be found in model.py.

Requirements

TensorFlow needs to be installed before running the training script. Code is tested on TensorFlow version 1.0.1 for Python 2.7 and Python 3.5.

In addition, librosa must be installed for reading and writing audio.

To install the required python packages, run

pip install -r requirements.txt

For GPU support, use

pip install -r requirements_gpu.txt

Training the network

You can use any corpus containing .wav files. We've mainly used the VCTK corpus (around 10.4GB, Alternative host) so far.

In order to train the network, execute

python train.py --data_dir=corpus

to train the network, where corpus is a directory containing .wav files. The script will recursively collect all .wav files in the directory.

You can see documentation on each of the training settings by running

python train.py --help

You can find the configuration of the model parameters in wavenet_params.json. These need to stay the same between training and generation.

Global Conditioning

Global conditioning refers to modifying the model such that the id of a set of mutually-exclusive categories is specified during training and generation of .wav file. In the case of the VCTK, this id is the integer id of the speaker, of which there are over a hundred. This allows (indeed requires) that a speaker id be specified at time of generation to select which of the speakers it should mimic. For more details see the paper or source code.

Training with Global Conditioning

The instructions above for training refer to training without global conditioning. To train with global conditioning, specify command-line arguments as follows:

python train.py --data_dir=corpus --gc_channels=32

The --gc_channels argument does two things:

  • It tells the train.py script that it should build a model that includes global conditioning.
  • It specifies the size of the embedding vector that is looked up based on the id of the speaker.

The global conditioning logic in train.py and audio_reader.py is "hard-wired" to the VCTK corpus at the moment in that it expects to be able to determine the speaker id from the pattern of file naming used in VCTK, but can be easily be modified.

Generating audio

Example output generated by @jyegerlehner based on speaker 280 from the VCTK corpus.

You can use the generate.py script to generate audio using a previously trained model.

Generating without Global Conditioning

Run

python generate.py --samples 16000 logdir/train/2017-02-13T16-45-34/model.ckpt-80000

where logdir/train/2017-02-13T16-45-34/model.ckpt-80000 needs to be a path to previously saved model (without extension). The --samples parameter specifies how many audio samples you would like to generate (16000 corresponds to 1 second by default).

The generated waveform can be played back using TensorBoard, or stored as a .wav file by using the --wav_out_path parameter:

python generate.py --wav_out_path=generated.wav --samples 16000 logdir/train/2017-02-13T16-45-34/model.ckpt-80000

Passing --save_every in addition to --wav_out_path will save the in-progress wav file every n samples.

python generate.py --wav_out_path=generated.wav --save_every 2000 --samples 16000 logdir/train/2017-02-13T16-45-34/model.ckpt-80000

Fast generation is enabled by default. It uses the implementation from the Fast Wavenet repository. You can follow the link for an explanation of how it works. This reduces the time needed to generate samples to a few minutes.

To disable fast generation:

python generate.py --samples 16000 logdir/train/2017-02-13T16-45-34/model.ckpt-80000 --fast_generation=false

Generating with Global Conditioning

Generate from a model incorporating global conditioning as follows:

python generate.py --samples 16000  --wav_out_path speaker311.wav --gc_channels=32 --gc_cardinality=377 --gc_id=311 logdir/train/2017-02-13T16-45-34/model.ckpt-80000

Where:

--gc_channels=32 specifies 32 is the size of the embedding vector, and must match what was specified when training.

--gc_cardinality=377 is required as 376 is the largest id of a speaker in the VCTK corpus. If some other corpus is used, then this number should match what is automatically determined and printed out by the train.py script at training time.

--gc_id=311 specifies the id of speaker, speaker 311, for which a sample is to be generated.

Running tests

Install the test requirements

pip install -r requirements_test.txt

Run the test suite

./ci/test.sh

Missing features

Currently there is no local conditioning on extra information which would allow context stacks or controlling what speech is generated.

Related projects

Owner
Igor Babuschkin
Igor Babuschkin
High-performance moving least squares material point method (MLS-MPM) solver.

High-Performance MLS-MPM Solver with Cutting and Coupling (CPIC) (MIT License) A Moving Least Squares Material Point Method with Displacement Disconti

Yuanming Hu 2.2k Dec 31, 2022
DiffQ performs differentiable quantization using pseudo quantization noise. It can automatically tune the number of bits used per weight or group of weights, in order to achieve a given trade-off between model size and accuracy.

Differentiable Model Compression via Pseudo Quantization Noise DiffQ performs differentiable quantization using pseudo quantization noise. It can auto

Facebook Research 145 Dec 30, 2022
Deep Learning Interviews book: Hundreds of fully solved job interview questions from a wide range of key topics in AI.

This book was written for you: an aspiring data scientist with a quantitative background, facing down the gauntlet of the interview process in an increasingly competitive field. For most of you, the

4.1k Dec 28, 2022
Python implementation of Project Fluent

Project Fluent This is a collection of Python packages to use the Fluent localization system. python-fluent consists of these packages: fluent.syntax

Project Fluent 155 Dec 28, 2022
Codes for our IJCAI21 paper: Dialogue Discourse-Aware Graph Model and Data Augmentation for Meeting Summarization

DDAMS This is the pytorch code for our IJCAI 2021 paper Dialogue Discourse-Aware Graph Model and Data Augmentation for Meeting Summarization [Arxiv Pr

xcfeng 55 Dec 27, 2022
Keras-1D-ACGAN-Data-Augmentation

Keras-1D-ACGAN-Data-Augmentation What is the ACGAN(Auxiliary Classifier GANs) ? Related Paper : [Abstract : Synthesizing high resolution photorealisti

Jae-Hoon Shim 7 Dec 23, 2022
A treasure chest for visual recognition powered by PaddlePaddle

简体中文 | English PaddleClas 简介 飞桨图像识别套件PaddleClas是飞桨为工业界和学术界所准备的一个图像识别任务的工具集,助力使用者训练出更好的视觉模型和应用落地。 近期更新 2021.11.1 发布PP-ShiTu技术报告,新增饮料识别demo 2021.10.23 发

4.6k Dec 31, 2022
joint detection and semantic segmentation, based on ultralytics/yolov5,

Multi YOLO V5——Detection and Semantic Segmentation Overeview This is my undergraduate graduation project which based on ultralytics YOLO V5 tag v5.0.

477 Jan 06, 2023
TART - A PyTorch implementation for Transition Matrix Representation of Trees with Transposed Convolutions

TART This project is a PyTorch implementation for Transition Matrix Representati

Lee Sael 2 Jan 19, 2022
CALVIN - A benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks

CALVIN CALVIN - A benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks Oier Mees, Lukas Hermann, Erick Rosete,

Oier Mees 107 Dec 26, 2022
Contrastive Learning for Compact Single Image Dehazing, CVPR2021

AECR-Net Contrastive Learning for Compact Single Image Dehazing, CVPR2021. Official Pytorch based implementation. Paper arxiv Pytorch Version TODO: mo

glassy 253 Jan 01, 2023
Pytorch implementation of AREL

Status: Archive (code is provided as-is, no updates expected) Agent-Temporal Attention for Reward Redistribution in Episodic Multi-Agent Reinforcement

8 Nov 25, 2022
Semantic Segmentation for Aerial Imagery using Convolutional Neural Network

This repo has been deprecated because whole things are re-implemented by using Chainer and I did refactoring for many codes. So please check this newe

Shunta Saito 27 Sep 23, 2022
Demos of essentia classifiers hosted on replicate.ai

essentia-replicate-demos Demos of Essentia models hosted on replicate.ai's MTG site. The models Check our site for a complete list of the models avail

Music Technology Group - Universitat Pompeu Fabra 12 Nov 14, 2022
Repo for paper "Dynamic Placement of Rapidly Deployable Mobile Sensor Robots Using Machine Learning and Expected Value of Information"

Repo for paper "Dynamic Placement of Rapidly Deployable Mobile Sensor Robots Using Machine Learning and Expected Value of Information" Notes I probabl

Berkeley Expert System Technologies Lab 0 Jul 01, 2021
This is a package for LiDARTag, described in paper: LiDARTag: A Real-Time Fiducial Tag System for Point Clouds

LiDARTag Overview This is a package for LiDARTag, described in paper: LiDARTag: A Real-Time Fiducial Tag System for Point Clouds (PDF)(arXiv). This wo

University of Michigan Dynamic Legged Locomotion Robotics Lab 159 Dec 21, 2022
Denoising Diffusion Implicit Models

Denoising Diffusion Implicit Models (DDIM) Jiaming Song, Chenlin Meng and Stefano Ermon, Stanford Implements sampling from an implicit model that is t

465 Jan 05, 2023
Music Source Separation; Train & Eval & Inference piplines and pretrained models we used for 2021 ISMIR MDX Challenge.

Introduction 1. Usage (For MSS) 1.1 Prepare running environment 1.2 Use pretrained model 1.3 Train new MSS models from scratch 1.3.1 How to train 1.3.

Leo 100 Dec 25, 2022
GLNet for Memory-Efficient Segmentation of Ultra-High Resolution Images

GLNet for Memory-Efficient Segmentation of Ultra-High Resolution Images Collaborative Global-Local Networks for Memory-Efficient Segmentation of Ultra-

VITA 298 Dec 12, 2022
A Low Complexity Speech Enhancement Framework for Full-Band Audio (48kHz) based on Deep Filtering.

DeepFilterNet A Low Complexity Speech Enhancement Framework for Full-Band Audio (48kHz) based on Deep Filtering. libDF contains Rust code used for dat

Hendrik Schröter 292 Dec 25, 2022