Towards Developing State-of-the-Art TTS Synthesisers for 13 Indian Languages with Signal Processing aided Alignments

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Prakash, Anusha, Umesh, S, Murthy, Hema A
Natura: Preprint
Pubblicazione: 2022
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910608413163520
author Prakash, Anusha
Umesh, S
Murthy, Hema A
author_facet Prakash, Anusha
Umesh, S
Murthy, Hema A
contents End-to-end (E2E) systems synthesise high-quality speech, but this typically requires a large amount of data. As E2E synthesis progressed from Tacotron to FastSpeech2, it became evident that features representing prosody, particularly sub-word durations, are important for error-free synthesis. Variants of FastSpeech use a teacher model or forced alignments for training. This paper uses signal processing cues in tandem with forced alignment to produce accurate phone boundaries for the training data. As a result of better duration modelling, good-quality synthesisers are developed. Evaluations indicate that systems developed using the proposed signal processing-aided approach are better than systems developed using other alignment approaches, especially in low-resource scenarios. Our systems also outperform the existing best TTS systems available for 13 Indian languages.
format Preprint
id arxiv_https___arxiv_org_abs_2210_17153
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Towards Developing State-of-the-Art TTS Synthesisers for 13 Indian Languages with Signal Processing aided Alignments
Prakash, Anusha
Umesh, S
Murthy, Hema A
Audio and Speech Processing
Sound
End-to-end (E2E) systems synthesise high-quality speech, but this typically requires a large amount of data. As E2E synthesis progressed from Tacotron to FastSpeech2, it became evident that features representing prosody, particularly sub-word durations, are important for error-free synthesis. Variants of FastSpeech use a teacher model or forced alignments for training. This paper uses signal processing cues in tandem with forced alignment to produce accurate phone boundaries for the training data. As a result of better duration modelling, good-quality synthesisers are developed. Evaluations indicate that systems developed using the proposed signal processing-aided approach are better than systems developed using other alignment approaches, especially in low-resource scenarios. Our systems also outperform the existing best TTS systems available for 13 Indian languages.
title Towards Developing State-of-the-Art TTS Synthesisers for 13 Indian Languages with Signal Processing aided Alignments
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2210.17153