English

Property-Aware Multi-Speaker Data Simulation: A Probabilistic Modelling Technique for Synthetic Data Generation

Audio and Speech Processing 2023-10-20 v1 Sound

Abstract

We introduce a sophisticated multi-speaker speech data simulator, specifically engineered to generate multi-speaker speech recordings. A notable feature of this simulator is its capacity to modulate the distribution of silence and overlap via the adjustment of statistical parameters. This capability offers a tailored training environment for developing neural models suited for speaker diarization and voice activity detection. The acquisition of substantial datasets for speaker diarization often presents a significant challenge, particularly in multi-speaker scenarios. Furthermore, the precise time stamp annotation of speech data is a critical factor for training both speaker diarization and voice activity detection. Our proposed multi-speaker simulator tackles these problems by generating large-scale audio mixtures that maintain statistical properties closely aligned with the input parameters. We demonstrate that the proposed multi-speaker simulator generates audio mixtures with statistical properties that closely align with the input parameters derived from real-world statistics. Additionally, we present the effectiveness of speaker diarization and voice activity detection models, which have been trained exclusively on the generated simulated datasets.

Keywords

Cite

@article{arxiv.2310.12371,
  title  = {Property-Aware Multi-Speaker Data Simulation: A Probabilistic Modelling Technique for Synthetic Data Generation},
  author = {Tae Jin Park and He Huang and Coleman Hooper and Nithin Koluguri and Kunal Dhawan and Ante Jukic and Jagadeesh Balam and Boris Ginsburg},
  journal= {arXiv preprint arXiv:2310.12371},
  year   = {2023}
}
R2 v1 2026-06-28T12:54:59.883Z