# Supervised and Unsupervised Approaches for Controlling Narrow Lexical Focus in Sequence-to-Sequence Speech Synthesis

arXiv: https://arxiv.org/abs/2101.09940 · HTML: https://mlsanity.com/paper/2101.09940/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Jan 2021.

## Built on by (most cited first, 8 of 8)

- [A Survey on Neural Speech Synthesis](https://mlsanity.com/paper/2106.15561/) (2021)
- [Hierarchical prosody modeling and control in non-autoregressive parallel neural TTS](https://mlsanity.com/paper/2110.02952/) (2021)
- [Emphasis control for parallel neural TTS](https://mlsanity.com/paper/2110.03012/) (2021)
- [Speak While You Think: Streaming Speech Synthesis During Text Generation](https://mlsanity.com/paper/2309.11210/) (2023)
- [EE-TTS: Emphatic Expressive TTS with Linguistic Information](https://mlsanity.com/paper/2305.12107/) (2023)
- [LaughNet: synthesizing laughter utterances from waveform silhouettes and a single laughter example](https://mlsanity.com/paper/2110.04946/) (2021)
- [Improving French Synthetic Speech Quality via SSML Prosody Control](https://mlsanity.com/paper/2508.17494/) (2025)
- [Emphasis Sensitivity in Speech Representations](https://mlsanity.com/paper/2508.11566/) (2025)

## Abstract (arXiv)

Although Sequence-to-Sequence (S2S) architectures have become state-of-the-art in speech synthesis, capable of generating outputs that approach the perceptual quality of natural samples, they are limited by a lack of flexibility when it comes to controlling the output. In this work we present a framework capable of controlling the prosodic output via a set of concise, interpretable, disentangled parameters. We apply this framework to the realization of emphatic lexical focus, proposing a variety of architectures designed to exploit different levels of supervision based on the availability of labeled resources. We evaluate these approaches via listening tests that demonstrate we are able to successfully realize controllable focus while maintaining the same, or higher, naturalness over an established baseline, and we explore how the different approaches compare when synthesizing in a target voice with or without labeled data.