POWSM: A Phonetic Open Whisper-Style Speech Foundation Model

TL;DR

POWSM is a unified speech model capable of performing multiple phonetic tasks such as ASR, G2P, and P2G, outperforming specialized models and enabling versatile speech processing.

Contribution

Introduces POWSM, the first unified framework for multiple phonetic tasks, facilitating seamless conversion between audio, text, and phones.

Findings

Outperforms or matches specialized PR models of similar size

Supports multiple tasks including G2P, P2G, and ASR

Openly released training data, code, and models

Abstract

Recent advances in spoken language processing have led to substantial progress in phonetic tasks such as automatic speech recognition (ASR), phone recognition (PR), grapheme-to-phoneme conversion (G2P), and phoneme-to-grapheme conversion (P2G). Despite their conceptual similarity, these tasks have largely been studied in isolation, each relying on task-specific architectures and datasets. In this paper, we introduce POWSM (Phonetic Open Whisper-style Speech Model), the first unified framework capable of jointly performing multiple phone-related tasks. POWSM enables seamless conversion between audio, text (graphemes), and phones, opening up new possibilities for universal and low-resource speech processing. Our model outperforms or matches specialized PR models of similar size (Wav2Vec2Phoneme and ZIPA) while jointly supporting G2P, P2G, and ASR. Our training data, code and models are released to foster open science.