# Single channel voice separation for unknown number of speakers under reverberant and noisy settings

arXiv: https://arxiv.org/abs/2011.02329 · HTML: https://mlsanity.com/paper/2011.02329/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Nov 2020.

## Built on by (most cited first, 10 of 10)

- [High Fidelity Neural Audio Compression](https://mlsanity.com/paper/2210.13438/) (2022)
- [Augmentation Invariant Discrete Representation for Generative Spoken Language Modeling](https://mlsanity.com/paper/2209.15483/) (2022)
- [ReVISE: Self-Supervised Speech Resynthesis with Visual Input for Universal and Generalized Speech Enhancement](https://mlsanity.com/paper/2212.11377/) (2022)
- [A Systematic Comparison of Phonetic Aware Techniques for Speech Enhancement](https://mlsanity.com/paper/2206.11000/) (2022)
- [Learning to Separate Voices by Spatial Regions](https://mlsanity.com/paper/2207.04203/) (2022)
- [Deep Audio Waveform Prior](https://mlsanity.com/paper/2207.10441/) (2022)
- [Single microphone speaker extraction using unified time-frequency Siamese-Unet](https://mlsanity.com/paper/2203.02941/) (2022)
- [Spiking Music: Audio Compression with Event Based Auto-encoders](https://mlsanity.com/paper/2402.01571/) (2024)
- [Online Self-Attentive Gated RNNs for Real-Time Speaker Separation](https://mlsanity.com/paper/2106.13493/) (2021)
- [Coarse-to-Fine Recursive Speech Separation for Unknown Number of Speakers](https://mlsanity.com/paper/2203.16054/) (2022)

## Abstract (arXiv)

We present a unified network for voice separation of an unknown number of speakers. The proposed approach is composed of several separation heads optimized together with a speaker classification branch. The separation is carried out in the time domain, together with parameter sharing between all separation heads. The classification branch estimates the number of speakers while each head is specialized in separating a different number of speakers. We evaluate the proposed model under both clean and noisy reverberant set-tings. Results suggest that the proposed approach is superior to the baseline model by a significant margin. Additionally, we present a new noisy and reverberant dataset of up to five different speakers speaking simultaneously.