How does machine understand the audio?

Tóm tắt báo cáo

Besides visual data or text data, audio data is everywhere around us and contains lots of valuable information. In the world, there are several success stories about applying AI in processing this modality to create fantastic products such as Amazon Alexa, Google Assistant, Shazam, Youtube, etc. Such products leverage audio information in speech recognition, speaker identification, audio classification, and so on. In this seminar, we will share the way machines understand audio data. Particularly, we will revise the origin of physical sound and how to convert the sound into a digital signal which can be understood by a machine.