# Maximizing Mutual Information for Tacotron

arXiv: https://arxiv.org/abs/1909.01145 · HTML: https://mlsanity.com/paper/1909.01145/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Aug 2019.

## Built on by (most cited first, 5 of 5)

- [A Survey on Neural Speech Synthesis](https://mlsanity.com/paper/2106.15561/) (2021)
- [Review of end-to-end speech synthesis technology based on deep learning](https://mlsanity.com/paper/2104.09995/) (2021)
- [Using multiple reference audios and style embedding constraints for speech synthesis](https://mlsanity.com/paper/2110.04451/) (2021)
- [Nana-HDR: A Non-attentive Non-autoregressive Hybrid Model for TTS](https://mlsanity.com/paper/2109.13673/) (2021)
- [Facial Landmark Predictions with Applications to Metaverse](https://mlsanity.com/paper/2209.14698/) (2022)

## Abstract (arXiv)

End-to-end speech synthesis methods already achieve close-to-human quality performance. However compared to HMM-based and NN-based frame-to-frame regression methods, they are prone to some synthesis errors, such as missing or repeating words and incomplete synthesis. We attribute the comparatively high utterance error rate to the local information preference of conditional autoregressive models, and the ill-posed training objective of the model, which describes mostly the training status of the autoregressive module, but rarely that of the condition module. Inspired by InfoGAN, we propose to maximize the mutual information between the text condition and the predicted acoustic features to strengthen the dependency between them for CAR speech synthesis model, which would alleviate the local information preference issue and reduce the utterance error rate. The training objective of maximizing mutual information can be considered as a metric of the dependency between the autoregressive module and the condition module. Experiment results show that our method can reduce the utterance error rate.