Speech Recognition and Spoken Language Understanding for Mobile Personal Assistants: A Case Study of "Shabette Concier"
Kosuke Tsujino, Shinya Iizuka, Yusuke Nakashima, Yoshinori Isoda · 2013
Recent success of mobile personal assistants deeply relies on the advancement in statistical automatic speech recognition (ASR) and spoken language understanding (SLU) technologies. This article describes practical design of ASR and SLU systems for "Shabette Concier", a voice-based personal assistant application commercially released by NTT docomo in Japan in March 2012. Utilization of a large amount of field speech data gathered from the actual service is focused and the potential of big data is discussed.