Beyond Oversmoothing: Evaluating DDPM and MSE for Scalable Speech Synthesis in ASR

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Minixhofer, Christoph, Klejch, Ondrej, Bell, Peter
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912074396860416
author Minixhofer, Christoph
Klejch, Ondrej
Bell, Peter
author_facet Minixhofer, Christoph
Klejch, Ondrej
Bell, Peter
contents Synthetically generated speech has rapidly approached human levels of naturalness. However, the paradox remains that ASR systems, when trained on TTS output that is judged as natural by humans, continue to perform badly on real speech. In this work, we explore whether this phenomenon is due to the oversmoothing behaviour of models commonly used in TTS, with a particular focus on the behaviour of TTS-for-ASR as the amount of TTS training data is scaled up. We systematically compare Denoising Diffusion Probabilistic Models (DDPM) to Mean Squared Error (MSE) based models for TTS, when used for ASR model training. We test the scalability of the two approaches, varying both the number hours, and the number of different speakers. We find that for a given model size, DDPM can make better use of more data, and a more diverse set of speakers, than MSE models. We achieve the best reported ratio between real and synthetic speech WER to date (1.46), but also find that a large gap remains.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12279
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Beyond Oversmoothing: Evaluating DDPM and MSE for Scalable Speech Synthesis in ASR
Minixhofer, Christoph
Klejch, Ondrej
Bell, Peter
Audio and Speech Processing
Artificial Intelligence
Computation and Language
Synthetically generated speech has rapidly approached human levels of naturalness. However, the paradox remains that ASR systems, when trained on TTS output that is judged as natural by humans, continue to perform badly on real speech. In this work, we explore whether this phenomenon is due to the oversmoothing behaviour of models commonly used in TTS, with a particular focus on the behaviour of TTS-for-ASR as the amount of TTS training data is scaled up. We systematically compare Denoising Diffusion Probabilistic Models (DDPM) to Mean Squared Error (MSE) based models for TTS, when used for ASR model training. We test the scalability of the two approaches, varying both the number hours, and the number of different speakers. We find that for a given model size, DDPM can make better use of more data, and a more diverse set of speakers, than MSE models. We achieve the best reported ratio between real and synthetic speech WER to date (1.46), but also find that a large gap remains.
title Beyond Oversmoothing: Evaluating DDPM and MSE for Scalable Speech Synthesis in ASR
topic Audio and Speech Processing
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2410.12279