Personalization of CTC-based End-to-End Speech Recognition Using Pronunciation-Driven Subword Tokenization

Open in new window