Learning to Maximize Speech Quality Directly Using MOS Prediction for Neural Text-to-Speech

• URL: http://arxiv.org/abs/2011.01174v5
• Date: Wed, 25 May 2022 07:07:13 GMT
• Title: Learning to Maximize Speech Quality Directly Using MOS Prediction for Neural Text-to-Speech
• Authors: Yeunju Choi, Youngmoon Jung, Youngjoo Suh, Hoirin Kim
• Abstract summary: We propose a novel method to improve speech quality by training a TTS model under the supervision of perceptual loss. We first pre-train a mean opinion score (MOS) prediction model and then train a TTS model to maximize the MOS of synthesized speech. The proposed method can be applied independently regardless of the TTS model architecture or the cause of speech quality degradation.
• Score: 15.796199345773873
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Although recent neural text-to-speech (TTS) systems have achieved high-quality speech synthesis, there are cases where a TTS system generates low-quality speech, mainly caused by limited training data or information loss during knowledge distillation. Therefore, we propose a novel method to improve speech quality by training a TTS model under the supervision of perceptual loss, which measures the distance between the maximum possible speech quality score and the predicted one. We first pre-train a mean opinion score (MOS) prediction model and then train a TTS model to maximize the MOS of synthesized speech using the pre-trained MOS prediction model. The proposed method can be applied independently regardless of the TTS model architecture or the cause of speech quality degradation and efficiently without increasing the inference time or model complexity. The evaluation results for the MOS and phone error rate demonstrate that our proposed approach improves previous models in terms of both naturalness and intelligibility.

Related papers

• EM-TTS: Efficiently Trained Low-Resource Mongolian Lightweight Text-to-Speech [4.91849983180793]
  We propose a lightweight Text-to-Speech (TTS) system based on deep convolutional neural networks. Our model consists of two stages: Text2Spectrum and SSRN. Experiments show that our model can reduce the training time and parameters while ensuring the quality and naturalness of the synthesized speech.
  arXiv  Detail & Related papers  (2024-03-13T01:27:57Z)
• Investigating Content-Aware Neural Text-To-Speech MOS Prediction Using Prosodic and Linguistic Features [54.48824266041105]
  Current state-of-the-art methods for automatic synthetic speech evaluation are based on MOS prediction neural models. We propose to include prosodic and linguistic features as additional inputs in MOS prediction systems. All MOS prediction systems are trained on SOMOS, a neural TTS-only dataset with crowdsourced naturalness MOS evaluations.
  arXiv  Detail & Related papers  (2022-11-01T09:18:50Z)
• Contextual-Utterance Training for Automatic Speech Recognition [65.4571135368178]
  We propose a contextual-utterance training technique which makes use of the previous and future contextual utterances. Also, we propose a dual-mode contextual-utterance training technique for streaming automatic speech recognition (ASR) systems. The proposed technique is able to reduce both the WER and the average last token emission latency by more than 6% and 40ms relative.
  arXiv  Detail & Related papers  (2022-10-27T08:10:44Z)
• Semi-Supervised Learning Based on Reference Model for Low-resource TTS [32.731900584216724]
  We propose a semi-supervised learning method for neural TTS in which labeled target data is limited. Experimental results show that our proposed semi-supervised learning scheme with limited target data significantly improves the voice quality for test data to achieve naturalness and robustness in speech synthesis.
  arXiv  Detail & Related papers  (2022-10-25T07:48:07Z)
• NaturalSpeech: End-to-End Text to Speech Synthesis with Human-Level Quality [123.97136358092585]
  We develop a TTS system called NaturalSpeech that achieves human-level quality on a benchmark dataset. Specifically, we leverage a variational autoencoder (VAE) for end-to-end text to waveform generation. Experiment evaluations on popular LJSpeech dataset show that our proposed NaturalSpeech achieves -0.01 CMOS to human recordings at the sentence level.
  arXiv  Detail & Related papers  (2022-05-09T16:57:35Z)
• SOMOS: The Samsung Open MOS Dataset for the Evaluation of Neural Text-to-Speech Synthesis [50.236929707024245]
  The SOMOS dataset is the first large-scale mean opinion scores (MOS) dataset consisting of solely neural text-to-speech (TTS) samples. It consists of 20K synthetic utterances of the LJ Speech voice, a public domain speech dataset.
  arXiv  Detail & Related papers  (2022-04-06T18:45:20Z)
• Deep Learning Based Assessment of Synthetic Speech Naturalness [14.463987018380468]
  We present a new objective prediction model for synthetic speech naturalness. It can be used to evaluate Text-To-Speech or Voice Conversion systems.
  arXiv  Detail & Related papers  (2021-04-23T16:05:20Z)
• Hierarchical Prosody Modeling for Non-Autoregressive Speech Synthesis [76.39883780990489]
  We analyze the behavior of non-autoregressive TTS models under different prosody-modeling settings. We propose a hierarchical architecture, in which the prediction of phoneme-level prosody features are conditioned on the word-level prosody features.
  arXiv  Detail & Related papers  (2020-11-12T16:16:41Z)
• Pretraining Techniques for Sequence-to-Sequence Voice Conversion [57.65753150356411]
  Sequence-to-sequence (seq2seq) voice conversion (VC) models are attractive owing to their ability to convert prosody. We propose to transfer knowledge from other speech processing tasks where large-scale corpora are easily available, typically text-to-speech (TTS) and automatic speech recognition (ASR) We argue that VC models with such pretrained ASR or TTS model parameters can generate effective hidden representations for high-fidelity, highly intelligible converted speech.
  arXiv  Detail & Related papers  (2020-08-07T11:02:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.