MBNET: MOS Prediction for Synthesized Speech with Mean-Bias Network

By Yichong Leng, Xu Tan, Sheng Zhao, Frank K. Soong, Xiang-Yang Li, Tao Qin. 2021

132 citations recorded by the source corpus.

Mean opinion score (MOS) is a popular subjective metric to assess the quality of synthesized speech, and usually involves multiple human judges to evaluate each speech utterance. To reduce the labor cost in MOS test, multiple methods have been proposed to automatically predict MOS scores. To our knowledge, for a speech utterance, all previous works only used the average of multiple scores from different judges as the training target and discarded the score of each individual judge, which did not well exploit the precious MOS training data. In this paper, we propose MBNet, a MOS predictor with a mean subnet and a bias subnet to better utilize every judge score in MOS datasets, where the mean subnet is used to predict the mean score of each utterance similar to that in previous works, and the bias subnet to predict the bias score (the difference between the mean score and each individual judge score) and capture the personal preference of individual judges. Experiments show that compared with MOSNet baseline that only leverages mean score for training, MBNet improves the system-level spearmans rank correlation coefficient (SRCC) by 2.9% on VCC 2018 dataset and 6.7% on VCC 2016 dataset.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xu Tan
2. Tao Qin
3. Tie-Yan Liu
4. Sheng Zhao
5. Lei He
6. Frank K. Soong
7. Yanqing Liu
8. Yichong Leng

In Signal Processing

1. Xu Tan
2. Tao Qin
3. Tie-Yan Liu
4. Lei He
5. Sheng Zhao
6. Frank K. Soong
7. Yanqing Liu
8. Hung-yi Lee

Citations in the corpus

• Meta-TTS: Meta-Learning for Few-Shot Speaker Adaptive Text-to-Speech (2021, background): “The training target of MOSNet [27] is to predict the average MOS of each utterance, while MBNet [28] also considers the judging bias of each rater.”
• NaturalSpeech: End-to-End Text-to-Speech Synthesis With Human-Level Quality (2022, background): “1) Judgment Guideline: While there are some objective metrics to measure the quality gap between the generated speech and human recordings, such as PESQ [30], STOI [31], SI-SDR [32], they are not reliable to measure the perception quality in TTS [33], [34].”
• Utilizing Self-supervised Representations for MOS Prediction (2021, methodology): “Utilizing individual judge’s scores in training has been shown to improve the performance [11].”
• DiffGAN-TTS: High-Fidelity and Efficient Text-to-Speech with Denoising Diffusion GANs (2022, unknown): “, MBNet(Leng et al., 2021), MOS predictor with Wav2vec2.”
• UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022 (2022, unknown): “In addition to the contrastive loss, we use the clipped MSE loss [4] for the regression loss between the discrete predicted and ground truth scores: L(y, ŷ) = 1(|y− ŷ| > τ)(y− ŷ)(2).”