音響・韻律摂動による音声品質評価における人間とモデルの乖離の調査
Investigating Human-Model Discrepancies in Speech Quality Assessment via Acoustic and Prosodic Perturbations
音声品質評価モデルが音響劣化には追従するが、韻律誤差や話者特性の変化に対する感度が人間と異なることを、制御された摂動実験で明らかにした論文。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Masato Takagi, Masaya Kawamura, Reo Shimizu, Yuma Shirahata
分類: eess.AS, cs.CL, cs.LG, cs.SD
原文アブストラクト
Mean opinion score (MOS) prediction models are widely used as proxy metrics in text-to-speech (TTS) research, yet their ability to capture quality differences beyond acoustic fidelity remains unclear. We investigate this via controlled perturbations on speech: acoustic degradation, prosodic errors, and manipulation of speaker-specific characteristics such as pitch and speaking rate. We obtained MOS predictions for these speech samples from both human listeners and the model, and analyzed the differences in their perceptual characteristics. Results show that most models track acoustic degradation well, while all are insensitive to prosodic errors despite large subjective score drops. For speaker characteristics, models exhibit a double dissociation: strong mean fundamental frequency (F0) biases absent in human ratings, yet insensitivity to speaking rate and F0 variability that humans notice. These findings highlight limitations of scalar MOS prediction beyond acoustic fidelity.
