STYLER: Style Factor Modeling with Rapidity and Robustness via Speech Decomposition for Expressive and Controllable Neural Text to Speech

Overview

Hits

STYLER: Style Factor Modeling with Rapidity and Robustness via Speech Decomposition for Expressive and Controllable Neural Text to Speech

Keon Lee, Kyumin Park, Daeyoung Kim

In our paper, we propose STYLER, a non-autoregressive TTS framework with style factor modeling that achieves rapidity, robustness, expressivity, and controllability at the same time.

Abstract: Previous works on neural text-to-speech (TTS) have been addressed on limited speed in training and inference time, robustness for difficult synthesis conditions, expressiveness, and controllability. Although several approaches resolve some limitations, there has been no attempt to solve all weaknesses at once. In this paper, we propose STYLER, an expressive and controllable TTS framework with high-speed and robust synthesis. Our novel audio-text aligning method called Mel Calibrator and excluding autoregressive decoding enable rapid training and inference and robust synthesis on unseen data. Also, disentangled style factor modeling under supervision enlarges the controllability in synthesizing process leading to expressive TTS. On top of it, a novel noise modeling pipeline using domain adversarial training and Residual Decoding empowers noise-robust style transfer, decomposing the noise without any additional label. Various experiments demonstrate that STYLER is more effective in speed and robustness than expressive TTS with autoregressive decoding and more expressive and controllable than reading style non-autoregressive TTS. Synthesis samples and experiment results are provided via our demo page, and code is available publicly.

Dependencies

Please install the python dependencies given in requirements.txt.

pip3 install -r requirements.txt

Training

Preparation

Clean Data

  1. Download VCTK dataset and resample audios to a 22050Hz sampling rate.
  2. We provide a bash script for the resampling. Refer to data/resample.sh for the detail.
  3. Put audio files and corresponding text (transcript) files in the same directory. Both audio and text files must have the same name, excluding the extension.
  4. You may need to trim the audio for stable model convergence. Refer to Yeongtae's preprocess_audio.py for helpful preprocessing, including the trimming.
  5. Modify the hp.data_dir in hparams.py.

Noisy Data

  1. Download WHAM! dataset and resample audios to a 22050Hz sampling rate.
  2. Modify the hp.noise_dir in hparams.py.

Vocoder

  1. Unzip hifigan/generator_universal.pth.tar.zip in the same directory.

Preprocess

First, download ResCNN Softmax+Triplet pretrained model of philipperemy's DeepSpeaker for the speaker embedding as described in our paper and locate it in hp.speaker_embedder_dir.

Second, download the Montreal Forced Aligner(MFA) package and the pretrained (LibriSpeech) lexicon file through the following commands. MFA is used to obtain the alignments between the utterances and the phoneme sequences as FastSpeech2.

wget https://github.com/MontrealCorpusTools/Montreal-Forced-Aligner/releases/download/v1.1.0-beta.2/montreal-forced-aligner_linux.tar.gz
tar -zxvf montreal-forced-aligner_linux.tar.gz

wget http://www.openslr.org/resources/11/librispeech-lexicon.txt -O montreal-forced-aligner/pretrained_models/librispeech-lexicon.txt

Then, process all the necessary features. You will get a stat.txt file in your hp.preprocessed_path/. You have to modify the f0 and energy parameters in the hparams.py according to the content of stat.txt.

python3 preprocess.py

Finally, get the noisy data separately from the clean data by mixing each utterance with a randomly selected piece of background noise from WHAM! dataset.

python3 preprocess_noisy.py

Train

Now you have all the prerequisites! Train the model using the following command:

python3 train.py

Inference

Prepare Texts

Create sentences.py in data/ which has a python list named sentences of texts to be synthesized. Note that sentences can contain more than one text.

# In 'data/sentences.py',
sentences = [
    "Nothing is lost, everything is recycled."
]

Prepare Reference Audios

Reference audio preparation has a similar process to training data preparation. There could be two kinds of references: clean and noisy.

First, put clean audios with corresponding texts in a single directory and modify the hp.ref_audio_dir in hparams.py and process all the necessary features. Refer to the Clean Data section of Train Preparation.

python3 preprocess_refs.py

Then, get the noisy references.

python3 preprocess_noisy.py --refs

Synthesize

The following command will synthesize all combinations of texts in data/sentences.py and audios in hp.ref_audio_dir.

python3 synthesize.py --ckpt CHECKPOINT_PATH

Or you can specify single reference audio in hp.ref_audio_dir as follows.

python3 synthesize.py --ckpt CHECKPOINT_PATH --ref_name AUDIO_FILENAME

Also, there are several useful options.

  1. --speaker_id will specify the speaker. The specified speaker's embedding should be in hp.preprocessed_path/spker_embed. The default value is None, and the speaker embedding is calculated at runtime on each input audio.

  2. --inspection will give you additional outputs that show the effects of each encoder of STYLER. The samples are the same as the Style Factor Modeling section on our demo page.

  3. --cont will generate the samples as the Style Factor Control section on our demo page.

    python3 synthesize.py --ckpt CHECKPOINT_PATH --cont --r1 AUDIO_FILENAME_1 --r2 AUDIO_FILENAME_1

    Note that --cont option is only working on preprocessed data. In detail, the audios' name should have the same format as VCTK dataset (e.g., p323_229), and the preprocessed data must be existing in hp.preprocessed_path.

TensorBoard

The TensorBoard loggers are stored in the log directory. Use

tensorboard --logdir log

to serve the TensorBoard on your localhost. Here are some logging views of the model training on VCTK for 560k steps.

Notes

  1. There were too many noise data where extraction was not possible through pyworld as in clean data. To resolve this, pysptk was applied to extract log f0 for the noisy data's fundamental frequency. The --noisy_input option will automate this process during synthesizing.

  2. If MFA-related problems occur during running preprocess.py, try to manually run MFA by the following command.

    # Replace $data_dir and $PREPROCESSED_PATH with ./VCTK-Corpus-92/wav48_silence_trimmed and ./preprocessed/VCTK/TextGrid, for example
    ./montreal-forced-aligner/bin/mfa_align $YOUR_data_dir montreal-forced-aligner/pretrained_models/librispeech-lexicon.txt english $YOUR_PREPROCESSED_PATH -j 8
  3. DeepSpeaker on VCTK dataset shows clear identification among speakers. The following figure shows the T-SNE plot of extracted speaker embedding in our experiments.

  4. Currently, preprocess.py divides the dataset into two subsets: train and validation set. If you need other sets, such as a test set, the only thing to do is modifying the text files (train.txt or val.txt) in hp.preprocessed_path/.

Citation

If you would like to use or refer to this implementation, please cite our paper with the repo.

@article{lee2021styler,
  title={STYLER: Style Modeling with Rapidity and Robustness via SpeechDecomposition for Expressive and Controllable Neural Text to Speech},
  author={Lee, Keon and Park, Kyumin and Kim, Daeyoung},
  journal={arXiv preprint arXiv:2103.09474},
  year={2021}
}

References

Comments
  • some questions

    some questions

    1.in paper, why output of encoder (text_encoding) upsample and downsample? 2. what is the meaning of text_encoding_neck+pitch_encoding、text_encoding_neck+energy_encoding? why not cat?

    opened by Pydataman 3
  • Low resource languages that won't work with MFA?

    Low resource languages that won't work with MFA?

    Is there a way to fine tune a model or training two languages side by side such that a very low resource language can be trained with the voices of a high resource language?

    opened by michael-conrad 3
  • Undefined names

    Undefined names

    Hi, I noticed some undefined names around the code:

    synthesize.py:495:67: F821 undefined name 'reference' noise_mixer_refs.py:56:42: F821 undefined name 'eps' noise_mixer_refs.py:59:40: F821 undefined name 'eps'

    opened by L3str4nge 2
  • About the pre-process

    About the pre-process

    Hi, I want to ask the trimming operation whether is very important for training your model? Furthermore, can you share the scripts to trimming VCTK dataset?

    opened by yangdongchao 0
  • Bump tensorflow from 2.4.0 to 2.5.1

    Bump tensorflow from 2.4.0 to 2.5.1

    Bumps tensorflow from 2.4.0 to 2.5.1.

    Release notes

    Sourced from tensorflow's releases.

    TensorFlow 2.5.1

    Release 2.5.1

    This release introduces several vulnerability fixes:

    • Fixes a heap out of bounds access in sparse reduction operations (CVE-2021-37635)
    • Fixes a floating point exception in SparseDenseCwiseDiv (CVE-2021-37636)
    • Fixes a null pointer dereference in CompressElement (CVE-2021-37637)
    • Fixes a null pointer dereference in RaggedTensorToTensor (CVE-2021-37638)
    • Fixes a null pointer dereference and a heap OOB read arising from operations restoring tensors (CVE-2021-37639)
    • Fixes an integer division by 0 in sparse reshaping (CVE-2021-37640)
    • Fixes a division by 0 in ResourceScatterDiv (CVE-2021-37642)
    • Fixes a heap OOB in RaggedGather (CVE-2021-37641)
    • Fixes a std::abort raised from TensorListReserve (CVE-2021-37644)
    • Fixes a null pointer dereference in MatrixDiagPartOp (CVE-2021-37643)
    • Fixes an integer overflow due to conversion to unsigned (CVE-2021-37645)
    • Fixes a bad allocation error in StringNGrams caused by integer conversion (CVE-2021-37646)
    • Fixes a null pointer dereference in SparseTensorSliceDataset (CVE-2021-37647)
    • Fixes an incorrect validation of SaveV2 inputs (CVE-2021-37648)
    • Fixes a null pointer dereference in UncompressElement (CVE-2021-37649)
    • Fixes a segfault and a heap buffer overflow in {Experimental,}DatasetToTFRecord (CVE-2021-37650)
    • Fixes a heap buffer overflow in FractionalAvgPoolGrad (CVE-2021-37651)
    • Fixes a use after free in boosted trees creation (CVE-2021-37652)
    • Fixes a division by 0 in ResourceGather (CVE-2021-37653)
    • Fixes a heap OOB and a CHECK fail in ResourceGather (CVE-2021-37654)
    • Fixes a heap OOB in ResourceScatterUpdate (CVE-2021-37655)
    • Fixes an undefined behavior arising from reference binding to nullptr in RaggedTensorToSparse (CVE-2021-37656)
    • Fixes an undefined behavior arising from reference binding to nullptr in MatrixDiagV* ops (CVE-2021-37657)
    • Fixes an undefined behavior arising from reference binding to nullptr in MatrixSetDiagV* ops (CVE-2021-37658)
    • Fixes an undefined behavior arising from reference binding to nullptr and heap OOB in binary cwise ops (CVE-2021-37659)
    • Fixes a division by 0 in inplace operations (CVE-2021-37660)
    • Fixes a crash caused by integer conversion to unsigned (CVE-2021-37661)
    • Fixes an undefined behavior arising from reference binding to nullptr in boosted trees (CVE-2021-37662)
    • Fixes a heap OOB in boosted trees (CVE-2021-37664)
    • Fixes vulnerabilities arising from incomplete validation in QuantizeV2 (CVE-2021-37663)
    • Fixes vulnerabilities arising from incomplete validation in MKL requantization (CVE-2021-37665)
    • Fixes an undefined behavior arising from reference binding to nullptr in RaggedTensorToVariant (CVE-2021-37666)
    • Fixes an undefined behavior arising from reference binding to nullptr in unicode encoding (CVE-2021-37667)
    • Fixes an FPE in tf.raw_ops.UnravelIndex (CVE-2021-37668)
    • Fixes a crash in NMS ops caused by integer conversion to unsigned (CVE-2021-37669)
    • Fixes a heap OOB in UpperBound and LowerBound (CVE-2021-37670)
    • Fixes an undefined behavior arising from reference binding to nullptr in map operations (CVE-2021-37671)
    • Fixes a heap OOB in SdcaOptimizerV2 (CVE-2021-37672)
    • Fixes a CHECK-fail in MapStage (CVE-2021-37673)
    • Fixes a vulnerability arising from incomplete validation in MaxPoolGrad (CVE-2021-37674)
    • Fixes an undefined behavior arising from reference binding to nullptr in shape inference (CVE-2021-37676)
    • Fixes a division by 0 in most convolution operators (CVE-2021-37675)
    • Fixes vulnerabilities arising from missing validation in shape inference for Dequantize (CVE-2021-37677)
    • Fixes an arbitrary code execution due to YAML deserialization (CVE-2021-37678)
    • Fixes a heap OOB in nested tf.map_fn with RaggedTensors (CVE-2021-37679)

    ... (truncated)

    Changelog

    Sourced from tensorflow's changelog.

    Release 2.5.1

    This release introduces several vulnerability fixes:

    • Fixes a heap out of bounds access in sparse reduction operations (CVE-2021-37635)
    • Fixes a floating point exception in SparseDenseCwiseDiv (CVE-2021-37636)
    • Fixes a null pointer dereference in CompressElement (CVE-2021-37637)
    • Fixes a null pointer dereference in RaggedTensorToTensor (CVE-2021-37638)
    • Fixes a null pointer dereference and a heap OOB read arising from operations restoring tensors (CVE-2021-37639)
    • Fixes an integer division by 0 in sparse reshaping (CVE-2021-37640)
    • Fixes a division by 0 in ResourceScatterDiv (CVE-2021-37642)
    • Fixes a heap OOB in RaggedGather (CVE-2021-37641)
    • Fixes a std::abort raised from TensorListReserve (CVE-2021-37644)
    • Fixes a null pointer dereference in MatrixDiagPartOp (CVE-2021-37643)
    • Fixes an integer overflow due to conversion to unsigned (CVE-2021-37645)
    • Fixes a bad allocation error in StringNGrams caused by integer conversion (CVE-2021-37646)
    • Fixes a null pointer dereference in SparseTensorSliceDataset (CVE-2021-37647)
    • Fixes an incorrect validation of SaveV2 inputs (CVE-2021-37648)
    • Fixes a null pointer dereference in UncompressElement (CVE-2021-37649)
    • Fixes a segfault and a heap buffer overflow in {Experimental,}DatasetToTFRecord (CVE-2021-37650)
    • Fixes a heap buffer overflow in FractionalAvgPoolGrad (CVE-2021-37651)
    • Fixes a use after free in boosted trees creation (CVE-2021-37652)
    • Fixes a division by 0 in ResourceGather (CVE-2021-37653)
    • Fixes a heap OOB and a CHECK fail in ResourceGather (CVE-2021-37654)
    • Fixes a heap OOB in ResourceScatterUpdate (CVE-2021-37655)
    • Fixes an undefined behavior arising from reference binding to nullptr in RaggedTensorToSparse

    ... (truncated)

    Commits
    • 8222c1c Merge pull request #51381 from tensorflow/mm-fix-r2.5-build
    • d584260 Disable broken/flaky test
    • f6c6ce3 Merge pull request #51367 from tensorflow-jenkins/version-numbers-2.5.1-17468
    • 3ca7812 Update version numbers to 2.5.1
    • 4fdf683 Merge pull request #51361 from tensorflow/mm-update-relnotes-on-r2.5
    • 05fc01a Put CVE numbers for fixes in parentheses
    • bee1dc4 Update release notes for the new patch release
    • 47beb4c Merge pull request #50597 from kruglov-dmitry/v2.5.0-sync-abseil-cmake-bazel
    • 6f39597 Merge pull request #49383 from ashahab/abin-load-segfault-r2.5
    • 0539b34 Merge pull request #48979 from liufengdb/r2.5-cherrypick
    • Additional commits viewable in compare view

    Dependabot compatibility score

    Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.


    Dependabot commands and options

    You can trigger Dependabot actions by commenting on this PR:

    • @dependabot rebase will rebase this PR
    • @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
    • @dependabot merge will merge this PR after your CI passes on it
    • @dependabot squash and merge will squash and merge this PR after your CI passes on it
    • @dependabot cancel merge will cancel a previously requested merge and block automerging
    • @dependabot reopen will reopen this PR if it is closed
    • @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
    • @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
    • @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
    • @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
    • @dependabot use these labels will set the current labels as the default for future PRs for this repo and language
    • @dependabot use these reviewers will set the current reviewers as the default for future PRs for this repo and language
    • @dependabot use these assignees will set the current assignees as the default for future PRs for this repo and language
    • @dependabot use this milestone will set the current milestone as the default for future PRs for this repo and language

    You can disable automated security fix PRs for this repo from the Security Alerts page.

    dependencies 
    opened by dependabot[bot] 0
Releases(v1.0.0)
Owner
Keon Lee
Expressive Speech Synthesis | Disentangled Representation | Generative Models | NLP | HCI
Keon Lee
:hot_pepper: R²SQL: "Dynamic Hybrid Relation Network for Cross-Domain Context-Dependent Semantic Parsing." (AAAI 2021)

R²SQL The PyTorch implementation of paper Dynamic Hybrid Relation Network for Cross-Domain Context-Dependent Semantic Parsing. (AAAI 2021) Requirement

huybery 60 Dec 31, 2022
Data and codes for ACL 2021 paper: Towards Emotional Support Dialog Systems

Emotional-Support-Conversation Copyright © 2021 CoAI Group, Tsinghua University. All rights reserved. Data and codes are for academic research use onl

126 Dec 21, 2022
[ACM MM 2021] TSA-Net: Tube Self-Attention Network for Action Quality Assessment

Tube Self-Attention Network (TSA-Net) This repository contains the PyTorch implementation for paper TSA-Net: Tube Self-Attention Network for Action Qu

ShunliWang 18 Dec 23, 2022
Turning SymPy expressions into PyTorch modules.

sympytorch A micro-library as a convenience for turning SymPy expressions into PyTorch Modules. All SymPy floats become trainable parameters. All SymP

Patrick Kidger 89 Dec 13, 2022
This repository contains the reference implementation for our proposed Convolutional CRFs.

ConvCRF This repository contains the reference implementation for our proposed Convolutional CRFs in PyTorch (Tensorflow planned). The two main entry-

Marvin Teichmann 553 Dec 07, 2022
A simple algorithm for extracting tree height in sparse scene from point cloud data.

TREE HEIGHT EXTRACTION IN SPARSE SCENES BASED ON UAV REMOTE SENSING This is the offical python implementation of the paper "Tree Height Extraction in

6 Oct 28, 2022
Anonymous implementation of KSL

k-Step Latent (KSL) Implementation of k-Step Latent (KSL) in PyTorch. Representation Learning for Data-Efficient Reinforcement Learning [Paper] Code i

1 Nov 10, 2021
🐥A PyTorch implementation of OpenAI's finetuned transformer language model with a script to import the weights pre-trained by OpenAI

PyTorch implementation of OpenAI's Finetuned Transformer Language Model This is a PyTorch implementation of the TensorFlow code provided with OpenAI's

Hugging Face 1.4k Jan 05, 2023
PG2Net: Personalized and Group PreferenceGuided Network for Next Place Prediction

PG2Net PG2Net:Personalized and Group Preference Guided Network for Next Place Prediction Datasets Experiment results on two Foursquare check-in datase

Urban Mobility 5 Dec 20, 2022
Official code release for "GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis"

GRAF This repository contains official code for the paper GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis. You can find detailed usage i

349 Dec 29, 2022
DLFlow is a deep learning framework.

DLFlow是一套深度学习pipeline,它结合了Spark的大规模特征处理能力和Tensorflow模型构建能力。利用DLFlow可以快速处理原始特征、训练模型并进行大规模分布式预测,十分适合离线环境下的生产任务。利用DLFlow,用户只需专注于模型开发,而无需关心原始特征处理、pipeline构建、生产部署等工作。

DiDi 152 Oct 27, 2022
Unofficial implementation of Fast-SCNN: Fast Semantic Segmentation Network

Fast-SCNN: Fast Semantic Segmentation Network Unofficial implementation of the model architecture of Fast-SCNN. Real-time Semantic Segmentation and mo

Philip Popien 69 Aug 11, 2022
A 35mm camera, based on the Canonet G-III QL17 rangefinder, simulated in Python.

c is for Camera A 35mm camera, based on the Canonet G-III QL17 rangefinder, simulated in Python. The purpose of this project is to explore and underst

Daniele Procida 146 Sep 26, 2022
Adaptive Prototype Learning and Allocation for Few-Shot Segmentation (CVPR 2021)

ASGNet The code is for the paper "Adaptive Prototype Learning and Allocation for Few-Shot Segmentation" (accepted to CVPR 2021) [arxiv] Overview data/

Gen Li 91 Dec 23, 2022
A tensorflow=1.13 implementation of Deconvolutional Networks on Graph Data (NeurIPS 2021)

GDN A tensorflow=1.13 implementation of Deconvolutional Networks on Graph Data (NeurIPS 2021) Abstract In this paper, we consider an inverse problem i

4 Sep 13, 2022
DenseCLIP: Language-Guided Dense Prediction with Context-Aware Prompting

DenseCLIP: Language-Guided Dense Prediction with Context-Aware Prompting Created by Yongming Rao*, Wenliang Zhao*, Guangyi Chen, Yansong Tang, Zheng Z

Yongming Rao 322 Dec 31, 2022
Pixel-wise segmentation on VOC2012 dataset using pytorch.

PiWiSe Pixel-wise segmentation on the VOC2012 dataset using pytorch. FCN SegNet PSPNet UNet RefineNet For a more complete implementation of segmentati

Bodo Kaiser 378 Dec 30, 2022
Wind Speed Prediction using LSTMs in PyTorch

Implementation of Deep-Forecast using PyTorch Deep Forecast: Deep Learning-based Spatio-Temporal Forecasting Adapted from original implementation Setu

Onur Kaplan 151 Dec 14, 2022
salabim - discrete event simulation in Python

Object oriented discrete event simulation and animation in Python. Includes process control features, resources, queues, monitors. statistical distrib

181 Dec 21, 2022
Official repository for "Orthogonal Projection Loss" (ICCV'21)

Orthogonal Projection Loss (ICCV'21) Kanchana Ranasinghe, Muzammal Naseer, Munawar Hayat, Salman Khan, & Fahad Shahbaz Khan Paper Link | Project Page

Kanchana Ranasinghe 83 Dec 26, 2022