# LASPA: Language Agnostic Speaker Disentanglement with Prefix-Tuned Cross-Attention

arXiv: https://arxiv.org/abs/2506.02083 · HTML: https://mlsanity.com/paper/2506.02083/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 32 references that are in the map.
- First posted on arXiv 2 Jun 2025.

## Builds on (most cited first, 21 of 21)

- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [Unsupervised Domain Adaptation by Backpropagation](https://mlsanity.com/paper/1409.7495/) (2014)
- [VoxCeleb2: Deep Speaker Recognition](https://mlsanity.com/paper/1806.05622/) (2018)
- [VoxCeleb: a large-scale speaker identification dataset](https://mlsanity.com/paper/1706.08612/) (2017)
- [ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification](https://mlsanity.com/paper/2005.07143/) (2020)
- [Towards Principled Methods for Training Generative Adversarial Networks](https://mlsanity.com/paper/1701.04862/) (2017)
- [In defence of metric learning for speaker recognition](https://mlsanity.com/paper/2003.11982/) (2020)
- [VoxLingua107: a Dataset for Spoken Language Recognition](https://mlsanity.com/paper/2011.12998/) (2020)
- [x-vectors meet emotions: A study on dependencies between emotion and speaker recognition](https://mlsanity.com/paper/2002.05039/) (2020)
- [Reshape Dimensions Network for Speaker Recognition](https://mlsanity.com/paper/2407.18223/) (2024)
- [The ins and outs of speaker recognition: lessons from VoxSRC 2020](https://mlsanity.com/paper/2010.15809/) (2020)
- [Probing the Information Encoded in X-vectors](https://mlsanity.com/paper/1909.06351/) (2019)
- [Accent and Speaker Disentanglement in Many-to-many Voice Conversion](https://mlsanity.com/paper/2011.08609/) (2020)
- [Disentangled Representation with Dual-stage Feature Learning for Face Anti-spoofing](https://mlsanity.com/paper/2110.09157/) (2021)
- [Cross-Modal Fusion and Attention Mechanism for Weakly Supervised Video Anomaly Detection](https://mlsanity.com/paper/2412.20455/) (2024)
- [Generating Multilingual Voices Using Speaker Space Translation Based on Bilingual Speaker Data](https://mlsanity.com/paper/2004.04972/) (2020)
- [NISP: A Multi-lingual Multi-accent Dataset for Speaker Profiling](https://mlsanity.com/paper/2007.06021/) (2020)
- [DISPLACE Challenge: DIarization of SPeaker and LAnguage in Conversational Environments](https://mlsanity.com/paper/2303.00830/) (2023)
- [Disentangled representation learning for multilingual speaker recognition](https://mlsanity.com/paper/2211.00437/) (2022)
- [Disentangled Speaker Representation Learning via Mutual Information Minimization](https://mlsanity.com/paper/2208.08012/) (2022)
- [Deep Representation Decomposition for Rate-Invariant Speaker Verification](https://mlsanity.com/paper/2205.14294/) (2022)

## Built on by (most cited first, 2 of 2)

- [L-Proto: Language-Aware Episodic Prototypical Training for Multilingual Speaker Verification](https://mlsanity.com/paper/2606.17416/) (2026)
- [Disentangling Speaker and Language Effects in Cross-Lingual Speaker Verification for Iberian Languages](https://mlsanity.com/paper/2607.01161/) (2026)

## Abstract (arXiv)

Speaker recognition models face challenges in multi-lingual settings due to the entanglement of linguistic information within speaker embeddings. The overlap between vocal traits such as accent, vocal anatomy, and a language's phonetic structure complicates separating linguistic and speaker information. Disentangling these components can significantly improve speaker recognition accuracy. To this end, we propose a novel disentanglement learning strategy that integrates joint learning through prefix-tuned cross-attention. This approach is particularly effective when speakers switch between languages. Experimental results show the model generalizes across monolingual and multi-lingual settings, including unseen languages. Notably, the proposed model improves the equal error rate across multiple datasets, highlighting its ability to separate language information from speaker embeddings and enhance recognition in diverse linguistic conditions.