Demographic Attributes Prediction from Speech Using WavLM Embeddings

TL;DR

This paper presents a WavLM-based classifier that accurately predicts demographic attributes from speech, improving performance over existing models and supporting applications like language learning and digital forensics.

Contribution

Introduces a novel demographic prediction framework using pretrained WavLM embeddings, achieving state-of-the-art accuracy and robustness across multiple datasets.

Findings

Achieves 4.94 MAE in age prediction

Over 99.81% accuracy in gender classification

Improves existing models by up to 30% in MAE

Abstract

This paper introduces a general classifier based on WavLM features, to infer demographic characteristics, such as age, gender, native language, education, and country, from speech. Demographic feature prediction plays a crucial role in applications like language learning, accessibility, and digital forensics, enabling more personalized and inclusive technologies. Leveraging pretrained models for embedding extraction, the proposed framework identifies key acoustic and linguistic fea-tures associated with demographic attributes, achieving a Mean Absolute Error (MAE) of 4.94 for age prediction and over 99.81% accuracy for gender classification across various datasets. Our system improves upon existing models by up to relative 30% in MAE and up to relative 10% in accuracy and F1 scores across tasks, leveraging a diverse range of datasets and large pretrained models to ensure robustness and generalizability. This study offers new insights into speaker diversity and provides a strong foundation for future research in speech-based demographic profiling.