Year: 2025
Type: conference-paper
Abstract: This paperproposes a web-based online voice chat robot, which supports speech recognition, speech synthesis, artificial intelligence and motion response with a 3D avatar.There are a lot of online chat robots which are based on Artificial Intelligence Markup Language (AIML), however, they usually only support text chat mode.Recently, a voice toolkit called MMDAgent is developed by Nagoya Institute of Technology, Japan, which supports speech recognition, synthesis and a 3D agent.However, the brain of MMDAgent is based on script instead of AIML and it could only support Japanese and run on local machine, which is not very convenient and intelligent.Therefore, we developed an online version voice chat robot called MMDAvatar, which also supports speech recognition, speech synthesis, and motion response with a 3D avatar.Furthermore, it is based on AIML and open to training online.Currently, for the speech recognition, we use the HTML5 speech recognition tag.The Festival speech synthesis engine is used to synthesize the speech.An AIML engine (program O) is used as the brain of the chat robot and the 3D avatar is drawn by WebGL.Compared to MMDAgent, with an AIML engine, our chat robot is more flexible.Meanwhile, since it is online and open to training, it will become more and more intelligent with the time.
Source: ASEE Midwest Section Conference
Authors: Jianming Liu, Steven L. Grant
Institution: Missouri University of Science and Technology
Language: English
Topic: Speech and dialogue systems
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4407701440