Voxlect: A Speech Foundation Model Benchmark for Modeling Dialects and Regional Languages Around the Globe

TL;DR

Voxlect is a comprehensive benchmark for evaluating speech foundation models' ability to classify dialects and regional languages globally, supporting diverse applications like ASR analysis and speech generation evaluation.

Contribution

The paper introduces Voxlect, a new benchmark with extensive dialectal speech data, enabling systematic evaluation of speech models across multiple languages and dialects.

Findings

Models show varying accuracy across dialects and languages.

Robustness of dialect classification under noisy conditions is assessed.

Voxlect enhances dialect-aware speech recognition and generation systems.

Abstract

We present Voxlect, a novel benchmark for modeling dialects and regional languages worldwide using speech foundation models. Specifically, we report comprehensive benchmark evaluations on dialects and regional language varieties in English, Arabic, Mandarin and Cantonese, Tibetan, Indic languages, Thai, Spanish, French, German, Brazilian Portuguese, and Italian. Our study used over 2 million training utterances from 30 publicly available speech corpora that are provided with dialectal information. We evaluate the performance of several widely used speech foundation models in classifying speech dialects. We assess the robustness of the dialectal models under noisy conditions and present an error analysis that highlights modeling results aligned with geographic continuity. In addition to benchmarking dialect classification, we demonstrate several downstream applications enabled by Voxlect. Specifically, we show that Voxlect can be applied to augment existing speech recognition datasets with dialect information, enabling a more detailed analysis of ASR performance across dialectal variations. Voxlect is also used as a tool to evaluate the performance of speech generation systems. Voxlect is publicly available with the license of the RAIL family at: https://github.com/tiantiaf0627/voxlect.