Recognition of vocal emotion in noise-vocoded speech by normal hearing and cochlear implant listeners

Zhi Zhu, Ryota Miyauchi, Yukiko Araki, Masashi Unoki · The Journal of the Acoustical Society of America · 2016

Chatterjee et al. (2015) reported that cochlear implant (CI) listeners have difficulty recognizing vocal emotions due to the limited spectral cues provided by CI devices. Researches on vocal emotion perception of CI listeners have been studying ways to simulate responses of CI listeners by using noise-vocoded speech (NVS) as CI simulations with normal-hearing (NH) listeners. However, it is still unclear whether the results of CI simulations with NH listeners are reliable with regards to CI listeners. This study aims to clarify whether CI listeners can perceive vocal emotion the same way as NH listeners with NVS do. Vocal-emotion recognition experiments were carried out by having both NH and CI listeners listen to original emotional speech and its NVS. The results for CI listeners revealed that they recognized sadness and hot anger more easily than joy and cold anger in both original emotional speech and NVS conditions. Moreover, the results for NH listeners with NVS showed the same trend. The results suggested that the vocal-emotion recognition paradigm using NVS can be used to investigate vocal emotion recognition by CI listeners. [Work supported by the Mitsubishi foundation and Grant in Aid for Scientific Research Innovative Areas (No. 16H01669) from MEXT, Japan.]

Read the paper · More papers on PaperTik