# Evaluating the Effectiveness of Pre-Trained Audio Embeddings for Classification of Parkinson's Disease Speech Data

arXiv: https://arxiv.org/abs/2506.02078 · HTML: https://mlsanity.com/paper/2506.02078/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 30 references that are in the map.
- First posted on arXiv 2 Jun 2025.

## Builds on (most cited first, 7 of 7)

- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [MLS: A Large-Scale Multilingual Dataset for Speech Research](https://mlsanity.com/paper/2012.03411/) (2020)
- [NeuroVoz: a Castillian Spanish corpus of parkinsonian speech](https://mlsanity.com/paper/2403.02371/) (2024)
- [Innovative Speech-Based Deep Learning Approaches for Parkinson's Disease Classification: A Systematic Review](https://mlsanity.com/paper/2407.17844/) (2024)
- [X-vectors: New Quantitative Biomarkers for Early Parkinson's Disease Detection from Speech](https://mlsanity.com/paper/2007.03599/) (2020)
- [Speech as a Biomarker for Disease Detection](https://mlsanity.com/paper/2409.10230/) (2024)
- [Audio-Based Deep Learning Frameworks for Detecting COVID-19](https://mlsanity.com/paper/2202.05626/) (2022)

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

- [A Benchmark for Early-stage Parkinson's Disease Detection from Speech](https://mlsanity.com/paper/2605.14066/) (2026)
- [Cross-Lingual Parkinson's Disease Severity Assessment Using Pre-trained Speech Embeddings: A Multi-Class Evaluation](https://mlsanity.com/paper/2609.20875/) (2026)

## Abstract (arXiv)

Speech impairments are prevalent biomarkers for Parkinson's Disease (PD), motivating the development of diagnostic techniques using speech data for clinical applications. Although deep acoustic features have shown promise for PD classification, their effectiveness often varies due to individual speaker differences, a factor that has not been thoroughly explored in the existing literature. This study investigates the effectiveness of three pre-trained audio embeddings (OpenL3, VGGish and Wav2Vec2.0 models) for PD classification. Using the NeuroVoz dataset, OpenL3 outperforms others in diadochokinesis (DDK) and listen and repeat (LR) tasks, capturing critical acoustic features for PD detection. Only Wav2Vec2.0 shows significant gender bias, achieving more favorable results for male speakers, in DDK tasks. The misclassified cases reveal challenges with atypical speech patterns, highlighting the need for improved feature extraction and model robustness in PD detection.