Comparison Of Speech Representations For Automatic Quality Estimation In Multi-speaker Text-to-speech Synthesis
2020 · Jennifer Williams, Joanna Rownicka, Pilar Oplustil, et al.
Abstract
We aim to characterize how different speakers contribute to the perceived output quality of multi-speaker Text-to-Speech (TTS) synthesis. We automatically rate the quality of TTS using a neural network (NN) trained on human mean opinion score (MOS) ratings. First, we train and evaluate our NN model on 13 different TTS and voice conversion (VC) systems from the ASVSpoof 2019 Logical Access (LA) Dataset. Since it is not known how best to represent speech for this task, we compare 8 different representations alongside MOSNet frame-based features. Our representations include image-based spectrogram features and x-vector embeddings that explicitly model different types of noise such as T60 reverberation time. Our NN predicts MOS with a high correlation to human judgments. We report prediction correlation and error. A key finding is the quality achieved for certain speakers seems consistent, regardless of the TTS or VC system. It is widely accepted that some speakers give higher quality than
Authors
(none)
Tags
Stats
Related papers
- A Comparison Of Deep Learning MOS Predictors For Speech Synthesis Quality (2022)6.34
- Automos: Learning A Non-intrusive Assessor Of Naturalness-of-speech (2016)0.00
- Non-intrusive Speech Quality Assessment Using Neural Networks (2019)13.74
- An Objective Evaluation Of The Effects Of Recording Conditions And Speaker Characteristics In Multi-speaker Deep Neural Speech Synthesis (2021)0.00
- Modeling Multi-speaker Latent Space To Improve Neural TTS: Quick Enrolling New Speaker And Enhancing Premium Voice (2018)0.00
- Learning To Maximize Speech Quality Directly Using MOS Prediction For Neural Text-to-speech (2020)7.81
- Combining Speakers Of Multiple Languages To Improve Quality Of Neural Voices (2021)5.24
- Naturalspeech: End-to-end Text To Speech Synthesis With Human-level Quality (2022)16.32