Fine-Grained and Interpretable Neural Speech Editing

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Morrison, Max, Churchwell, Cameron, Pruyne, Nathan, Pardo, Bryan
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866929413130551296
author Morrison, Max
Churchwell, Cameron
Pruyne, Nathan
Pardo, Bryan
author_facet Morrison, Max
Churchwell, Cameron
Pruyne, Nathan
Pardo, Bryan
contents Fine-grained editing of speech attributes$\unicode{x2014}$such as prosody (i.e., the pitch, loudness, and phoneme durations), pronunciation, speaker identity, and formants$\unicode{x2014}$is useful for fine-tuning and fixing imperfections in human and AI-generated speech recordings for creation of podcasts, film dialogue, and video game dialogue. Existing speech synthesis systems use representations that entangle two or more of these attributes, prohibiting their use in fine-grained, disentangled editing. In this paper, we demonstrate the first disentangled and interpretable representation of speech with comparable subjective and objective vocoding reconstruction accuracy to Mel spectrograms. Our interpretable representation, combined with our proposed data augmentation method, enables training an existing neural vocoder to perform fast, accurate, and high-quality editing of pitch, duration, volume, timbral correlates of volume, pronunciation, speaker identity, and spectral balance.
format Preprint
id arxiv_https___arxiv_org_abs_2407_05471
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Fine-Grained and Interpretable Neural Speech Editing
Morrison, Max
Churchwell, Cameron
Pruyne, Nathan
Pardo, Bryan
Audio and Speech Processing
Sound
Fine-grained editing of speech attributes$\unicode{x2014}$such as prosody (i.e., the pitch, loudness, and phoneme durations), pronunciation, speaker identity, and formants$\unicode{x2014}$is useful for fine-tuning and fixing imperfections in human and AI-generated speech recordings for creation of podcasts, film dialogue, and video game dialogue. Existing speech synthesis systems use representations that entangle two or more of these attributes, prohibiting their use in fine-grained, disentangled editing. In this paper, we demonstrate the first disentangled and interpretable representation of speech with comparable subjective and objective vocoding reconstruction accuracy to Mel spectrograms. Our interpretable representation, combined with our proposed data augmentation method, enables training an existing neural vocoder to perform fast, accurate, and high-quality editing of pitch, duration, volume, timbral correlates of volume, pronunciation, speaker identity, and spectral balance.
title Fine-Grained and Interpretable Neural Speech Editing
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2407.05471