First trial of Yamaha HEARTalk UU-002

Yamaha Corporation’s HEARTalk does not recognize speech, but determines mood from the voices’ strength, length, shortness, high and low, between, and intonation. It judges mood in real time. This time I tried HEARTalk with the experimental board UU-002. Actually, I bought this UU-002 over two years ago. Since there is a student who love this technology, I decided to post it.

On the left side of the photo is UU-002, on the right side is the speaker. The golden part immediately to the left of the green LED of UU-002 is a MEMS microphone. When speaking some phrases “Hey”, “How to do”, “Terrible!” and so on to the board, it answers with a female voice, “Yes” or “Eh?.” The emotional information judged from the sound input from the microphone is output from the terminal of this board, so if you connect with a microcomputer such as Raspberry Pi or Arduino, you can do even more advanced processing. It is also possible to make the sound file different.

I was planning to evaluate this UU-002 for experiments to extract emergency information from the quantification of urgency by voice analysis, but I enjoyed myself very much. Interesting technology.