# Effect of data reduction on sequence-to-sequence neural TTS

arXiv: https://arxiv.org/abs/1811.06315 · HTML: https://mlsanity.com/paper/1811.06315/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Nov 2018.

## Built on by (most cited first, 16 of 16)

- [Machine Translation Verbosity Control for Automatic Dubbing](https://mlsanity.com/paper/2110.03847/) (2021)
- [Non-Autoregressive TTS with Explicit Duration Modelling for Low-Resource Highly Expressive Speech](https://mlsanity.com/paper/2106.12896/) (2021)
- [Voice Filter: Few-shot text-to-speech speaker adaptation using voice conversion as a post-processing module](https://mlsanity.com/paper/2202.08164/) (2022)
- [Prosodic Alignment for off-screen automatic dubbing](https://mlsanity.com/paper/2204.02530/) (2022)
- [Distribution augmentation for low-resource expressive text-to-speech](https://mlsanity.com/paper/2202.06409/) (2022)
- [Unify and Conquer: How Phonetic Feature Representation Affects Polyglot Text-To-Speech (TTS)](https://mlsanity.com/paper/2207.01547/) (2022)
- [Computer-assisted Pronunciation Training -- Speech synthesis is almost all you need](https://mlsanity.com/paper/2207.00774/) (2022)
- [Mix and Match: An Empirical Study on Training Corpus Composition for Polyglot Text-To-Speech (TTS)](https://mlsanity.com/paper/2207.01507/) (2022)
- [Combining speakers of multiple languages to improve quality of neural voices](https://mlsanity.com/paper/2108.07737/) (2021)
- [Custom Data Augmentation for low resource ASR using Bark and Retrieval-Based Voice Conversion](https://mlsanity.com/paper/2311.14836/) (2023)
- [A multilingual training strategy for low resource Text to Speech](https://mlsanity.com/paper/2409.01217/) (2024)
- [Enhancing audio quality for expressive Neural Text-to-Speech](https://mlsanity.com/paper/2108.06270/) (2021)
- [LibriS2S: A German-English Speech-to-Speech Translation Corpus](https://mlsanity.com/paper/2204.10593/) (2022)
- [Building a Luganda Text-to-Speech Model From Crowdsourced Data](https://mlsanity.com/paper/2405.10211/) (2024)
- [Sprachsynthese -- State-of-the-Art in englischer und deutscher Sprache](https://mlsanity.com/paper/2106.06230/) (2021)
- [LuxEmo: Expressive Text-to-Speech Corpus for Luxembourgish](https://mlsanity.com/paper/2606.31947/) (2026)

## Abstract (arXiv)

Recent speech synthesis systems based on sampling from autoregressive neural networks models can generate speech almost undistinguishable from human recordings. However, these models require large amounts of data. This paper shows that the lack of data from one speaker can be compensated with data from other speakers. The naturalness of Tacotron2-like models trained on a blend of 5k utterances from 7 speakers is better than that of speaker dependent models trained on 15k utterances, but in terms of stability multi-speaker models are always more stable. We also demonstrate that models mixing only 1250 utterances from a target speaker with 5k utterances from another 6 speakers can produce significantly better quality than state-of-the-art DNN-guided unit selection systems trained on more than 10 times the data from the target speaker.