# SLM-SS: Speech Language Model for Generative Speech Separation

arXiv: https://arxiv.org/abs/2601.19533 · HTML: https://mlsanity.com/paper/2601.19533/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 26 references that are in the map.
- First posted on arXiv 27 Jan 2026.

## Builds on (most cited first, 24 of 24)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [High Fidelity Neural Audio Compression](https://mlsanity.com/paper/2210.13438/) (2022)
- [Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation](https://mlsanity.com/paper/1809.07454/) (2018)
- [Seed-TTS: A Family of High-Quality Versatile Speech Generation Models](https://mlsanity.com/paper/2406.02430/) (2024)
- [Permutation Invariant Training of Deep Models for Speaker-Independent Multi-talker Speech Separation](https://mlsanity.com/paper/1607.00325/) (2016)
- [Attention is All You Need in Speech Separation](https://mlsanity.com/paper/2010.13154/) (2020)
- [LibriMix: An Open-Source Dataset for Generalizable Speech Separation](https://mlsanity.com/paper/2005.11262/) (2020)
- [TasNet: time-domain audio separation network for real-time, single-channel speech separation](https://mlsanity.com/paper/1711.00541/) (2017)
- [Serialized Output Training for End-to-End Overlapped Speech Recognition](https://mlsanity.com/paper/2003.12687/) (2020)
- [Music Source Separation with Band-split RNN](https://mlsanity.com/paper/2209.15174/) (2022)
- [Seed-ASR: Understanding Diverse Speech and Contexts with LLM-based Speech Recognition](https://mlsanity.com/paper/2407.04675/) (2024)
- [FireRedASR: Open-Source Industrial-Grade Mandarin Speech Recognition Models from Encoder-Decoder to LLM Integration](https://mlsanity.com/paper/2501.14350/) (2025)
- [Diffusion-based Generative Speech Source Separation](https://mlsanity.com/paper/2210.17327/) (2022)
- [SpeechBERTScore: Reference-Aware Automatic Evaluation of Speech Generation Leveraging NLP Evaluation Metrics](https://mlsanity.com/paper/2401.16812/) (2024)
- [SELM: Speech Enhancement Using Discrete Tokens and Language Models](https://mlsanity.com/paper/2312.09747/) (2023)
- [Adapting Multi-Lingual ASR Models for Handling Multiple Talkers](https://mlsanity.com/paper/2305.18747/) (2023)
- [TSELM: Target Speaker Extraction using Discrete Tokens and Language Models](https://mlsanity.com/paper/2409.07841/) (2024)
- [Cross-Speaker Encoding Network for Multi-Talker Speech Recognition](https://mlsanity.com/paper/2401.04152/) (2024)
- [TokenSplit: Using Discrete Speech Representations for Direct, Refined, and Transcript-Conditioned Speech Separation and Recognition](https://mlsanity.com/paper/2308.10415/) (2023)
- [SepALM: Audio Language Models Are Error Correctors for Robust Speech Separation](https://mlsanity.com/paper/2505.03273/) (2025)
- [UniSep: Universal Target Audio Separation with Language Models at Scale](https://mlsanity.com/paper/2503.23762/) (2025)

## Abstract (arXiv)

Speech separation (SS) has advanced significantly with neural network-based methods, showing improved performance on signal-level metrics. However, these methods often struggle to maintain speech intelligibility in the separated signals, which can negatively affect the performance of downstream tasks such as speech recognition. In this work, we propose SLM-SS, a novel approach that applies speech language models to SS, aiming to enhance the intelligibility and coherence of the separated signals. We frame SS as discrete multi-codebook sequence generation, using Encoder-Decoder models to map quantized speech mixtures to target tokens. In addition to the autoregressive modeling strategy, we introduce a non-autoregressive model to improve decoding efficiency for residual tokens. Experimental results on the LibriMix dataset demonstrate that our approach shows significantly better preservation of speech intelligibility, leading to improved linguistic consistency in a variety of downstream tasks compared to existing approaches.