Multilingual Phonological Feature Recognition with Self-Supervised Speech Models
2026-05-25 • Computation and Language
Computation and Language
AI summaryⓘ
The authors developed PhonoQ-2.0, a tool that recognizes detailed speech features directly from audio in multiple languages without first identifying whole phonemes. Instead of guessing individual sounds, it predicts a set of 22 speech characteristics each moment, like how the sound is made or if it is voiced. They introduced a special method to make sure these features make sense together. Tests showed their system works better than traditional models both on known and new languages.
phonological featuresself-supervised learningspeech recognitionphonemeframe-level analysismanner of articulationvoicingCTC phoneme baselinemultilingual speech models
Authors
Abner Hernandez, Tomás Arias-Vergara, Daiqi Liu, Andreas Maier, Paula Andrea Pérez-Toro
Abstract
Phonological features provide a language-general and linguistically grounded representation of speech. We present PhonoQ-2.0, a multilingual frame-level phonological feature recognizer built on self-supervised speech models. The system directly predicts a structured 22-dimensional feature vector per frame encoding manner, vowel quality, place, and voicing, instead of deriving features from phoneme outputs. To ensure phonologically coherent predictions, we introduce a manner-conditioned gating mechanism that activates valid feature groups. Evaluated across multiple languages and corpora, PhonoQ-2.0 achieves an average macro-F1 of 91.3% in-domain and 88.9% out-of-domain. Compared to a strong CTC phoneme baseline, it delivers consistent gains of +8.8 F1 in-domain and +8.6 out-of-domain on average. In unseen-language evaluation, PhonoQ-2.0 improves macro-F1 from 66.9% to 73.6% (+6.7 on average), with gains of up to +10.8 points.