# DisfluencySpeech -- Single-Speaker Conversational Speech Dataset with Paralanguage

arXiv: https://arxiv.org/abs/2406.08820 · HTML: https://mlsanity.com/paper/2406.08820/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 23 references that are in the map.
- First posted on arXiv 13 Jun 2024.

## Builds on (most cited first, 10 of 10)

- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [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)
- [DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset](https://mlsanity.com/paper/1710.03957/) (2017)
- [Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Search](https://mlsanity.com/paper/2005.11129/) (2020)
- [NeMo: a toolkit for building AI applications using Neural Modules](https://mlsanity.com/paper/1909.09577/) (2019)
- [A Survey on Neural Speech Synthesis](https://mlsanity.com/paper/2106.15561/) (2021)
- [DailyTalk: Spoken Dialogue Dataset for Conversational Text-to-Speech](https://mlsanity.com/paper/2207.01063/) (2022)
- [One TTS Alignment To Rule Them All](https://mlsanity.com/paper/2108.10447/) (2021)
- [fairseq S^2: A Scalable and Integrable Speech Synthesis Toolkit](https://mlsanity.com/paper/2109.06912/) (2021)

## Built on by (most cited first, 6 of 6)

- [NVSpeech: An Integrated and Scalable Pipeline for Human-Like Speech Modeling with Paralinguistic Vocalizations](https://mlsanity.com/paper/2508.04195/) (2025)
- [NV-Bench: Benchmark of Nonverbal Vocalization Synthesis for Expressive Text-to-Speech Generation](https://mlsanity.com/paper/2603.15352/) (2026)
- [Transcription Policy as a Latent Variable: Activating Controllable Verbatim ASR with Word-Level Timing](https://mlsanity.com/paper/2607.18934/) (2026)
- [MUGEN: Evaluating and Improving Multi-audio Understanding of Large Audio-Language Models](https://mlsanity.com/paper/2603.09714/) (2026)
- [Preference Optimization for Non-Verbal Vocalization Synthesis](https://mlsanity.com/paper/2608.24163/) (2026)
- [ALARM: Audio-Language Alignment for Reasoning Models](https://mlsanity.com/paper/2603.09556/) (2026)

## Abstract (arXiv)

Laughing, sighing, stuttering, and other forms of paralanguage do not contribute any direct lexical meaning to speech, but they provide crucial propositional context that aids semantic and pragmatic processes such as irony. It is thus important for artificial social agents to both understand and be able to generate speech with semantically-important paralanguage. Most speech datasets do not include transcribed non-lexical speech sounds and disfluencies, while those that do are typically multi-speaker datasets where each speaker provides relatively little audio. This makes it challenging to train conversational Text-to-Speech (TTS) synthesis models that include such paralinguistic components. We thus present DisfluencySpeech, a studio-quality labeled English speech dataset with paralanguage. A single speaker recreates nearly 10 hours of expressive utterances from the Switchboard-1 Telephone Speech Corpus (Switchboard), simulating realistic informal conversations. To aid the development of a TTS model that is able to predictively synthesise paralanguage from text without such components, we provide three different transcripts at different levels of information removal (removal of non-speech events, removal of non-sentence elements, and removal of false starts), as well as benchmark TTS models trained on each of these levels.