# A Large-Scale Probing Analysis of Speaker-Specific Attributes in Self-Supervised Speech Representations

arXiv: https://arxiv.org/abs/2501.05310 · HTML: https://mlsanity.com/paper/2501.05310/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Jan 2025.

## Built on by (most cited first, 6 of 6)

- [Interpreting Speaker Characteristics in the Dimensions of Self-Supervised Speech Features](https://mlsanity.com/paper/2603.03096/) (2026)
- [Privacy-Preserving End-to-End Full-Duplex Speech Dialogue Models](https://mlsanity.com/paper/2603.08179/) (2026)
- [Beyond Decodability: Reconstructing Language Model Representations with an Encoding Probe](https://mlsanity.com/paper/2605.00607/) (2026)
- [Motor, Cognitive, or Corpus? What Survives Cross-Lingual Transfer in Speech-Based Parkinsons Disease Detection](https://mlsanity.com/paper/2608.13425/) (2026)
- [Kraken: LLM-based Speech-to-Speech Translation via Low-bitrate VQ and Dual-path Source Conditioning](https://mlsanity.com/paper/2609.13045/) (2026)
- [Feature Encoding in VAE-based Audio Decoders: Effects of Input, Depth and Distribution](https://mlsanity.com/paper/2610.07966/) (2026)

## Abstract (arXiv)

Enhancing explainability in speech self-supervised learning (SSL) is important for understanding and effectively utilising speech SSL representations. This study conducts a large-scale layer-wise probing analysis of 11 speech SSL models, examining speaker identity together with acoustic, prosodic, and paralinguistic attributes. The results confirm a general hierarchy wherein initial layers encode fundamental acoustics and middle layers synthesise abstract traits. The consensus that final layers purely abstract linguistic content is challenged, as larger models unexpectedly recover speaker-discriminative information in their deep layers. Furthermore, the intermediate representations of speech SSL models are found to provide stronger probing performance for several attributes than specialised speaker embeddings. These insights give a systematic empirical characterisation of speaker-related information in SSL models, providing guidelines for selecting representations for downstream tasks.