Does Listening Matter? Backchanneling and Nodding in AI Clone
Organizations: Kyoto University Kyoto, Japan · Sony Computer Science Laboratories Tokyo, Japan
Abstract
AI clones that imitate a specific person typically reproduce what the person says and how they sound, but not how they listen. We investigate whether adding multimodal listening behaviors gives such a clone more presence and authenticity. We integrated verbal backchannels and head nodding, driven by real-time prediction models, into an AI clone equipped with voice cloning and LLM-based responses. In a within-subjects study (N=35), adding these behaviors significantly improved the perceived attentiveness of the avatar, the sense of talking with the real person, and the feeling of co-presence. These results indicate that AI clone fidelity should extend beyond voice and response content to include interactive listening behavior.
Figures & tables
| Measure | With | Without | |
| User behavior | |||
| User utterances | 12.89 (2.52) | 13.11 (2.13) | .516 |
| Mean utterance length (s) | 0 7.30 (2.91) | 0 6.93 (3.15) | .195 |
| Dialogue duration (min) | 0 4.90 (0.25) | 0 4.83 (0.19) | .300 |
| System feedback behavior | |||
| Backchannels | 27.29 (7.55) | — | — |