Disentangling Correlated Speaker and Noise for Speech Synthesis via Data Augmentation and Adversarial Factorization
The DOI matches its Crossref record; 7 of the 7 references it cites were found in a registry or an index.
Abstract
To leverage crowd-sourced data to train multi-speaker text-to-speech (TTS) models that can synthesize clean speech for all speakers, it is essential to learn disentangled representations which can independently control the speaker identity and background noise in generated signals. However, learning such representations can be challenging, due to the lack of labels describing the recording conditions of each training example, and the fact that speakers and recording conditions are often correlated, e.g. since users often make many recordings using the same equipment. This paper proposes three components to address this problem by: (1) formulating a conditional generative model with factorized latent variables, (2) using data augmentation to add noise that is not correlated with speaker identity and whose label is known during training, and (3) using adversarial factorization to improve disentanglement. Experimental results demonstrate that the proposed method can disentangle speaker and noise attributes even if they are correlated in the training data, and can be used to consistently synthesize clean speech for all speakers. Ablation studies verify the importance of each proposed component.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Domain-Adversarial Training of Neural Networks
Ganin, Y., Ustinova, E., Ajakan, H., Germain, P., Larochelle, H., Laviolette, F., Marchand, M., & Lempitsky, V. (2017) · Advances in Computer Vision and Pattern Recognition · doi:10.1007/978-3-319-58347-1_10 · Cited by 8,318
“Librispeech: an ASR corpus based on public domain audio books,”
Panayotov, V., Chen, G., Povey, D., & Khudanpur, S. (2015) · 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2015.7178964 · Cited by 6,072
Analysis of Representations for Domain Adaptation
Ben-David, S., Blitzer, J., Crammer, K., & Pereira, F. (2007) · Advances in Neural Information Processing Systems 19 · doi:10.7551/mitpress/7503.003.0022 · Cited by 2,019 · cited 1×
“Following [3, 4], we approximate it with the Proxy A-distance: 2(1− 2 ), where is a generalization error of an augmentation classifier trained to predict if zs is inferred from an augmented utterance.”
MethodTacotron: towards end-to-end speech synthesis
Wang, Y., Skerry-Ryan, R. J., Stanton, D., Wu, Y., Weiss, R. J., Jaitly, N., Yang, Z., Xiao, Y., Chen, Z., Bengio, S., et al. (2017) · Interspeech 2017 · doi:10.21437/interspeech.2017-1452 · Cited by 1,746
Self-citationTransfer learning from speaker verification to multispeaker text-to-speech synthesis
Jia, Y., Zhang, Y., Weiss, R. J., Wang, Q., Shen, J., Ren, F., Chen, Z., Nguyen, P., Pang, R., Moreno, I. L., et al. (2018) · arXiv · doi:10.48550/arxiv.1806.04558 · Cited by 519 · cited 1×
“Recent development of neural end-to-end TTS models [27, 2] enables control of both labelled and unlabelled speech attributes by conditioning synthesis on both text and learned attribute representations [28, 22, 11, 1, 6, 10].”
BackgroundSelf-citationSUPERSEDED - CSTR VCTK Corpus: English Multi-speaker Corpus for CSTR Voice Cloning Toolkit
Veaux, C., Yamagishi, J., & MacDonald, K. (2016) · University of Edinburgh. The Centre for Speech Technology Research (CSTR) · doi:10.7488/ds/1495 · Cited by 434
Cited by
A review of deep learning techniques for speech processing
Mehrish, A., Majumder, N., Bharadwaj, R., Mihalcea, R., & Poria, S. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.101869 · Cited by 389
A survey on neural speech synthesis
Tan, X., Qin, T., Soong, F., & Liu, T.-Y. (2021) · arXiv · doi:10.48550/arxiv.2106.15561 · Cited by 216
“Since TTS is a typical sequence generation task and can be modeled through typical generative models, we can categorize in terms of different generative models: normal sequence generation model, flow, GAN, VAE, and diffusion model.”
BackgroundMethodInfluentialAttentron: Few-Shot Text-to-Speech Utilizing Attention-Based Variable-Length Embedding
Choi, S., Han, S., Kim, D., & Ha, S. (2020) · Interspeech 2020 · doi:10.21437/interspeech.2020-2096 · Cited by 56
SpeechNet: A Universal Modularized Model for Speech Processing Tasks
Chen, Y.-C., Chi, P.-H., Yang, S.-W., Chang, K.-W., Lin, J.-H., Huang, S.-F., Liu, D.-R., Liu, C.-L., Lee, C.-K., & Lee, H.-Y. (2021) · arXiv · doi:10.48550/arxiv.2105.03070 · Cited by 6
“SE and VC help SC, while TTS harm SC. SE helps TTS on both seen and unseen speakers.”
BackgroundMethodInfluential
Metadata
- Authors
- Hsu, W.-N., Zhang, Y., Weiss, R. J., Chung, Y.-A., Wang, Y., Wu, Y., & Glass, J.
- Venue
- ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Publisher
- IEEE
- Published
- 2019-05
- Pages
- 5901-5905
- ISSN
- 1520-6149 (ISSN-L)
- OpenAlex
- W2907262790
- Cited by
- 111
- References
- 7
- FWCI
- 8.31
- Type
- conference paper
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.