Intelligent Audio Processing Technology

Prototyping
Professor Name：Kuang-Yow Lian

In recent years, the laboratory has focused on the development of audio processing technology. We have built up several smart living systems using voice-related technology. For example, we can regulate building lighting, air conditioning, and electrical appliances using the keywords spoken by users. The core technologies include: (a) The unique noise reduction technology; (b) The microphone array for sound source orientation, sound source tracking, audio enhancement and multi-source identification technology; (c) Detection of baby sound and accent of different nationalities; (d) English accent correction system; (e) Ambient sound detection technology. In addition to the above-mentioned voice control technology constructed by intelligent algorithms, the laboratory also devotes to the artificial intelligence (AI) research, and develops high-precision automatic speech recognition (ASR) for keyword spotting using deep learning methods. The above innovative technologies are easily introduced into the application scenarios of smart medical care to improve the quality of medical life.
   The development of these technologies relies on well-developed high-performance algorithms. Some of the results have been highly recognized by the most famous practical competitions or well-known large companies.