# On the Role of Conversational Timing in Synthetic Training Data for ASR

arXiv: https://arxiv.org/abs/2607.08371 · HTML: https://mlsanity.com/paper/2607.08371/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 46 references that are in the map.
- First posted on arXiv 9 Jul 2026.

## Builds on (most cited first, 12 of 12)

- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [NeMo: a toolkit for building AI applications using Neural Modules](https://mlsanity.com/paper/1909.09577/) (2019)
- [CHiME-6 Challenge:Tackling Multispeaker Speech Recognition for Unsegmented Recordings](https://mlsanity.com/paper/2004.09249/) (2020)
- [Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition](https://mlsanity.com/paper/2305.05084/) (2023)
- [The fifth 'CHiME' Speech Separation and Recognition Challenge: Dataset, task and baselines](https://mlsanity.com/paper/1803.10609/) (2018)
- [End-to-End Neural Speaker Diarization with Permutation-Free Objectives](https://mlsanity.com/paper/1909.05952/) (2019)
- [Serialized Output Training for End-to-End Overlapped Speech Recognition](https://mlsanity.com/paper/2003.12687/) (2020)
- [From Simulated Mixtures to Simulated Conversations as Training Data for End-to-End Neural Diarization](https://mlsanity.com/paper/2204.00890/) (2022)
- [Improving the Naturalness of Simulated Conversations for End-to-End Neural Diarization](https://mlsanity.com/paper/2204.11232/) (2022)
- [Multi-Speaker and Wide-Band Simulated Conversations as Training Data for End-to-End Neural Diarization](https://mlsanity.com/paper/2211.06750/) (2022)
- [Toward Conversational Hungarian Speech Recognition: Introducing the BEA-Large and BEA-Dialogue Datasets](https://mlsanity.com/paper/2511.13529/) (2025)
- [Speaker-Aware Simulation Improves Conversational Speech Recognition](https://mlsanity.com/paper/2602.04776/) (2026)

## Abstract (arXiv)

Synthetic multi-speaker conversations are widely used to train conversational automatic speech recognition (ASR) systems, but it remains unclear which timing properties make simulated data most useful. This paper studies conversational timing as a controllable training variable rather than merely as a corpus statistic to be reproduced. We parameterize pause and overlap timing distributions with an exponential-tilting family estimated from multiple conversational corpora, and then explore the resulting four-dimensional parameter space with Latin hypercube sampling and multi-objective Bayesian optimization. Each sampled timing configuration is used to generate simulated training conversations, train an ASR system, and evaluate concatenated-permutation word and character error rates (cpWER and cpCER) on a Hungarian dialogue corpus. The results show that downstream ASR behavior is explained more directly by induced timing statistics than by raw simulator coordinates or corpus proximity. In particular, higher overlap exposure is associated with lower cpWER, whereas longer and more variable gaps are associated with higher cpWER; cpCER follows the same trend, but with weaker statistical support. Bayesian optimization yields modest aggregate improvements, but its main value is analytical: it produces controlled timing interventions that reveal an overlap--gap trade-off in simulated conversational training data. These findings suggest that realistic simulation should be complemented by task-relevant diagnostics of overlap, gap, and timing-variability profiles.