Montreal Forced Aligner and the state of speech-to-text alignment in 2026

2026-06-16

AI Summary

This study addresses the challenge of insufficient accuracy in speech-text forced alignment for low-resource and out-of-distribution languages by systematically upgrading the Montreal Forced Aligner to version 3.0. The proposed approach integrates an IPA-normalized pronunciation dictionary, probabilistic pronunciation modeling, phonological rule mechanisms, cross-lingual phoneme remapping, and model adaptation strategies within a hybrid evaluation framework that combines neural and traditional aligners. This enhancement significantly improves generalization across diverse linguistic contexts. Evaluated on four benchmark datasets—including English, Japanese, and Korean—the system achieves or approaches state-of-the-art performance, with an average boundary error below 15 milliseconds, demonstrating its effectiveness and robustness in multilingual scenarios.

Abstract

The Montreal Forced Aligner (MFA) was released in 2016 and has since become the most widely used tool for forced alignment in research and industry. In the decade since, MFA has undergone substantial development, including expanded coverage across more languages and dialects using larger open-source datasets, harmonized IPA dictionaries, model adaptation, cross-language phone remapping, and support utilities. This paper documents MFA 3.0's developments since version 1.0 and evaluates MFA's performance across English, Japanese, and Korean, benchmarked against classic and neural forced aligners. MFA 3.0 achieves state-of-the-art or near state-of-the-art performance across all four benchmark datasets with mean boundary errors below 15 ms. Adaptation and cross-language remapping are effective for languages outside MFA's training distribution, and pronunciation probability modeling and phonological rules provide gains in specific conditions.

Problem

• forced alignment
• speech-to-text alignment
• multilingual
• pronunciation modeling
• cross-language adaptation

Innovation

• forced alignment
• cross-language phone remapping
• model adaptation
• IPA dictionaries
• pronunciation probability modeling