# Towards Robust FastSpeech 2 by Modelling Residual Multimodality

arXiv: https://arxiv.org/abs/2306.01442 · HTML: https://mlsanity.com/paper/2306.01442/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 29 references that are in the map.
- First posted on arXiv 2 Jun 2023.

## Builds on (most cited first, 16 of 16)

- [WaveNet: A Generative Model for Raw Audio](https://mlsanity.com/paper/1609.03499/) (2016)
- [HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2010.05646/) (2020)
- [FastSpeech 2: Fast and High-Quality End-to-End Text to Speech](https://mlsanity.com/paper/2006.04558/) (2020)
- [LibriTTS: A Corpus Derived from LibriSpeech for Text-to-Speech](https://mlsanity.com/paper/1904.02882/) (2019)
- [Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech](https://mlsanity.com/paper/2106.06103/) (2021)
- [Tacotron: Towards End-to-End Speech Synthesis](https://mlsanity.com/paper/1703.10135/) (2017)
- [MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesis](https://mlsanity.com/paper/1910.06711/) (2019)
- [WaveGlow: A Flow-based Generative Network for Speech Synthesis](https://mlsanity.com/paper/1811.00002/) (2018)
- [Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Search](https://mlsanity.com/paper/2005.11129/) (2020)
- [Efficient Neural Audio Synthesis](https://mlsanity.com/paper/1802.08435/) (2018)
- [End-to-End Adversarial Text-to-Speech](https://mlsanity.com/paper/2006.03575/) (2020)
- [CDPAM: Contrastive learning for perceptual audio similarity](https://mlsanity.com/paper/2102.05109/) (2021)
- [Chunked Autoregressive GAN for Conditional Waveform Synthesis](https://mlsanity.com/paper/2110.10139/) (2021)
- [Investigating on Incorporating Pretrained and Learnable Speaker Representations for Multi-Speaker Multi-Style Text-to-Speech](https://mlsanity.com/paper/2103.04088/) (2021)
- [Revisiting Over-Smoothness in Text to Speech](https://mlsanity.com/paper/2202.13066/) (2022)
- [JETS: Jointly Training FastSpeech2 and HiFi-GAN for End to End Text to Speech](https://mlsanity.com/paper/2203.16852/) (2022)

## Built on by (most cited first, 2 of 2)

- [Acquiring Pronunciation Knowledge from Transcribed Speech Audio via Multi-task Learning](https://mlsanity.com/paper/2409.09891/) (2024)
- [A survey of AI-generated voices and their detection](https://mlsanity.com/paper/2608.15411/) (2026)

## Abstract (arXiv)

State-of-the-art non-autoregressive text-to-speech (TTS) models based on FastSpeech 2 can efficiently synthesise high-fidelity and natural speech. For expressive speech datasets however, we observe characteristic audio distortions. We demonstrate that such artefacts are introduced to the vocoder reconstruction by over-smooth mel-spectrogram predictions, which are induced by the choice of mean-squared-error (MSE) loss for training the mel-spectrogram decoder. With MSE loss FastSpeech 2 is limited to learn conditional averages of the training distribution, which might not lie close to a natural sample if the distribution still appears multimodal after all conditioning signals. To alleviate this problem, we introduce TVC-GMM, a mixture model of Trivariate-Chain Gaussian distributions, to model the residual multimodality. TVC-GMM reduces spectrogram smoothness and improves perceptual audio quality in particular for expressive datasets as shown by both objective and subjective evaluation.