mlsanityData to 7 Oct 2026

Distribution augmentation for low-resource expressive text-to-speech

arXiv 2202.06409 ↗First posted 13 Feb 2022eess.AS cs.CL cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

This paper presents a novel data augmentation technique for text-to-speech (TTS), that allows to generate new (text, audio) training examples without requiring any additional data. Our goal is to increase diversity of text conditionings available during training. This helps to reduce overfitting, especially in low-resource settings. Our method relies on substituting text and audio fragments in a way that preserves syntactical correctness. We take additional measures to ensure that synthesized speech does not contain artifacts caused by combining inconsistent audio samples. The perceptual evaluations show that our method improves speech quality over a number of datasets, speakers, and TTS architectures. We also demonstrate that it greatly improves robustness of attention-based TTS models.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 21 of its 27 references that are in the map. First posted on arXiv 13 Feb 2022.

Builds on

21 of 27 references.

References · 21

  1. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,261
  2. 2016WaveNet: A Generative Model for Raw Audiocited by 1,710
  3. 2017Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictionscited by 800
  4. 2015Attention-Based Models for Speech Recognitioncited by 400
  5. 2017The Effectiveness of Data Augmentation in Image Classification using Deep Learningcited by 368
  6. 2017Parallel WaveNet: Fast High-Fidelity Speech Synthesiscited by 214
  7. 2018Constituency Parsing with a Self-Attentive Encodercited by 196
  8. 2020Non-Attentive Tacotron: Robust and Controllable Neural TTS Synthesis Including Unsupervised Duration Modelingcited by 65
  9. 2021Hi-Fi Multi-Speaker English TTS Datasetcited by 60
  10. 2020Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesiscited by 54
  11. 2018Predicting Expressive Speaking Style From Text In End-To-End Speech Synthesiscited by 31
  12. 2018Semi-Supervised Training for Improving Data Efficiency in End-to-End Speech Synthesiscited by 30
9 more
  1. 2019End-to-end Text-to-speech for Low-resource Languages by Cross-Lingual Transfer Learningcited by 30
  2. 2020Low-resource expressive text-to-speech using data augmentationcited by 26
  3. 2021Substructure Substitution: Structured Data Augmentation for NLPcited by 25
  4. 2021Universal Neural Vocoding with Parallel WaveNetcited by 25
  5. 2020CopyCat: Many-to-Many Fine-Grained Prosody Transfer for Neural Text-to-Speechcited by 22
  6. 2018Effect of data reduction on sequence-to-sequence neural TTScited by 16
  7. 2021Non-Autoregressive TTS with Explicit Duration Modelling for Low-Resource Highly Expressive Speechcited by 14
  8. 2021Expressive Neural Voice Cloningcited by 11
  9. 2019Using VAEs and Normalizing Flows for One-shot Text-To-Speech Synthesis of Expressive Speechcited by 8

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 1; 2024: 2.

5 papers.

  1. 2024BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of datacited by 74
  2. 2022An Overview of Affective Speech Synthesis and Conversion in the Deep Learning Eracited by 32
  3. 2023Controllable Emphasis with zero data for text-to-speechcited by 4
  4. 2024ELAICHI: Enhancing Low-resource TTS by Addressing Infrequent and Low-frequency Character Bigramscited by 2
  5. 2022RWEN-TTS: Relation-aware Word Encoding Network for Natural Text-to-Speech Synthesiscited by 0