Emotion-based Music Recommender System for Smart Speaker
Autor: | LIN, HSIN-YUN, 林欣韻 |
---|---|
Rok vydání: | 2018 |
Druh dokumentu: | 學位論文 ; thesis |
Popis: | 106 Announced 2014, Amazon’s Echo connected with Alexa service, the hands-free voice control Echo had brought up the trend of VPA (Virtual Personal Assistant) speak-ers. There are numbers of applications had been developed for VPA speakers, and Music player is one of the most popular applications. VPA Speakers play music by clear voice command from Users. On the other hand, the mature technology of Music Emotion Recognition had been used on music playback as Emotion-based Music Recommender System, which is widely installed on other devices. This study proposes an Emotion-based Music Recommender System (MRS) em-bedded on VPA Speaker, which this pure voice interface system. All the communica-tions between VPA Speakers and users are short and voice commands are passive ac-ceptance, with limited user feedback. Emotion-based MRS of this study designed by different approach, first detect user’s location and local weather for preliminary estima-tion, afterward recommend proper music genre actively. With users’ feedback, Emo-tion-based MRS on VPA Speaker could learn to understand user’s preference. Emo-tion-based MRS of this study could be used on other voice interface systems, such as Robotics, Car audio system and mobile communication devices. |
Databáze: | Networked Digital Library of Theses & Dissertations |
Externí odkaz: |
načítá se...