4 Speech Recognition for Persian
Mahsa Vafaie, Jon Dehdari · 2023
Automatic Speech Recognition (ASR) is a cross-disciplinary field that enables computers to process human speech into written form and consequently make meaning of it with the help of other NLP technologies such as natural language understanding and natural language interpretation. Automatic speech recognition has benefited a great deal from advances in machine learning throughout the past decade. In this chapter a brief summary of ASR technologies and advances in Persian ASR is provided. Moreover, datasets and tools for Persian ASR are introduced and following a discussion of the peculiarities of Persian speech, some future directions and research gaps in the implementation of speech recognition technologies for Persian are addressed.