Automatic Pronunciation Generation by Utilizing a Semi-supervised Deep Neural Networks
This addresses pronunciation dictionary creation challenges for ASR systems, though it appears incremental as it builds on existing semi-supervised DNN approaches.
The paper tackles the problem of suboptimal phonemic units in ASR by proposing a data-driven method that jointly estimates sub-word units and dictionaries from orthographic transcriptions, showing it largely outperforms phoneme-based continuous speech recognition on the TIMIT dataset.
Phonemic or phonetic sub-word units are the most commonly used atomic elements to represent speech signals in modern ASRs. However they are not the optimal choice due to several reasons such as: large amount of effort required to handcraft a pronunciation dictionary, pronunciation variations, human mistakes and under-resourced dialects and languages. Here, we propose a data-driven pronunciation estimation and acoustic modeling method which only takes the orthographic transcription to jointly estimate a set of sub-word units and a reliable dictionary. Experimental results show that the proposed method which is based on semi-supervised training of a deep neural network largely outperforms phoneme based continuous speech recognition on the TIMIT dataset.