Yusuke Kanamori
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Yusuke Kanamori has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
Hiroshi Saruwatari
The University of Tokyo Japan
1 shared publicationShinnosuke Takamichi
The University of Tokyo Japan
1 shared publicationTaisei Takano
The University of Tokyo Japan
1 shared publication- Yuki Okamoto
The University of Tokyo Japan
1 shared publication Yuki Saito
The University of Tokyo Japan
1 shared publication
Latest Publications
- Open Access
RELATE: Subjective evaluation dataset for automatic evaluation of relevance between text and audio
In text-to-audio (TTA) research, the relevance between input text and output audio is an important evaluation aspect. Traditionally, it has been evaluated from both subjective and objective perspectives. However, subjective evaluation is costly in terms of money and time, and objective evaluation is unclear regarding the correlation to subjective evaluation scores. In this study, we construct RELATE, an open-sourced dataset that subjectively evaluates the relevance. Also, we benchmark a model for automatically predicting the subjective evaluation score from synthesized audio. Our model outperforms a conventional CLAPScore model, and that trend extends to many sound categories.
Open AccessRELATE: Subjective evaluation dataset for automatic evaluation of relevance between text and audio
In text-to-audio (TTA) research, the relevance between input text and output audio is an important evaluation aspect. Traditionally, it has been evaluated from both subjective and objective perspectives. However, subjective evaluation is costly in terms of money and time, and objective evaluation is unclear regarding the correlation to subjective evaluation scores. In this study, we construct RELATE, an open-sourced dataset that subjectively evaluates the relevance. Also, we benchmark a model for automatically predicting the subjective evaluation score from synthesized audio. Our model outperforms a conventional CLAPScore model, and that trend extends to many sound categories.