2017 · 43 citations · 14 references
ComputingComputer ScienceComputational LinguisticsNatural Language ProcessingPhonologyLanguage ModelingMachine TranslationMultilingual Natural Language ProcessingMultilingual ProcessingSpeech RecognitionAutomatic RecognitionGrapheme-phoneme ConversionGrapheme-phoneme MappingMultilingual Speech RecognitionMultilingual Text-to-speechPhoneme RecognitionSequence-to-sequence GenerationSequence-to-sequence Recognition
Grapheme-to-phoneme conversion is essential for text‑to‑speech and automatic speech recognition, yet most systems are monolingual and difficult to extend to low‑resource languages that lack data or handcrafted rules. We present a neural sequence‑to‑sequence approach to g2p trained on spelling–pronunciation pairs in hundreds of languages. The system shares a single encoder and decoder across all languages, enabling it to exploit intrinsic similarities between different writing systems. The model achieves an 11 % lower phoneme error rate than adapted high‑resource monolingual models and is more compact.
Grapheme-to-phoneme conversion (g2p) is necessary for text-to-speech and automatic speech recognition systems. Most g2p systems are monolingual: they require language-specific data or handcrafting of rules. Such systems are difficult to extend to low resource languages, for which data and handcrafted rules are not available. As an alternative, we present a neural sequence-to-sequence approach to g2p which is trained on spelling–pronunciation pairs in hundreds of languages. The system shares a single encoder and decoder across all languages, allowing it to utilize the intrinsic similarities between different writing systems. We show an 11% improvement in phoneme error rate over an approach based on adapting high-resource monolingual g2p models to low-resource languages. Our model is also much more compact relative to previous approaches.
14