Controllable Sequence-To-Sequence Neural TTS with LPCNET Backend for Real-time Speech Synthesis on CPU

Shechtman, Slava; Rabinovitz, Carmel; Sorin, Alex; Kons, Zvi; Hoory, Ron

doi:10.48550/arxiv.2002.10708

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When precise duration labels are unavailable or sparse in the training data, the proposed model can be trained explicitly with duration labels or unsupervised or semi-supervised using a fine-grained variational auto-encoder . When trained and wholly supervised, the proposed model slightly beats Tacotron 2 on naturalness [13]. The suggested model outperforms Tacotron 2 on naturalness with unsupervised or semi-supervised duration modelling while still much more resilient on over-generation and equivalent on under-generation.…”

mentioning

confidence: 85%

Deep Machine Learning-based Analysis for Intelligent Phonetic Language Recognition

Liu,

Luo

2024

SCPE

View full text Add to dashboard Cite

Modern speech generating systems can produce results that are almost as visually realistic as actual sounds. They still require further production management. This research presents a paradigm for managing prosodic output using explicit, unambiguous, and understandable parameters. We utilize this strategy to emphasize key words and provide a variety of architectural possibilities based on a richness of labelled resources. In an objective voice, we compare the options for producing data with or without labels. We assess them using listening tests that demonstrate our ability to retain the same level of naturalness while effectively attaining regulated concentration over a specific area.

show abstract

mentioning

confidence: 85%

Deep Machine Learning-based Analysis for Intelligent Phonetic Language Recognition

Liu,

Luo

2024

SCPE

View full text Add to dashboard Cite

show abstract

“…Those that require control use additional inputs to the networks as in Fig. 1a [4,5,2,6,7]. Those that offer control (but do not require it) aim at controlling latent variables while ensuring that they are interpretable [8,9,10,11], or disentangle one interpretable control variable (often the duration of the utterance) and make it controllable as in Fig.…”

Section: Related Workmentioning

confidence: 99%

Speech Synthesis and Control Using Differentiable DSP

Fabbro,

Golkov,

Kemp

et al. 2020

Preprint

View full text Add to dashboard Cite

Modern text-to-speech systems are able to produce natural and highquality speech, but speech contains factors of variation (e.g. pitch, rhythm, loudness, timbre) that text alone cannot contain. In this work we move towards a speech synthesis system that can produce diverse speech renditions of a text by allowing (but not requiring) explicit control over the various factors of variation. We propose a new neural vocoder that offers control of such factors of variation. This is achieved by employing differentiable digital signal processing (DDSP) (previously used only for music rather than speech), which exposes these factors of variation. The results show that the proposed approach can produce natural speech with realistic timbre, and individual factors of variation can be freely controlled.

show abstract

Controllable Sequence-To-Sequence Neural TTS with LPCNET Backend for Real-time Speech Synthesis on CPU

Cited by 2 publications

References 0 publications

Deep Machine Learning-based Analysis for Intelligent Phonetic Language Recognition

Deep Machine Learning-based Analysis for Intelligent Phonetic Language Recognition

Speech Synthesis and Control Using Differentiable DSP

Contact Info

Product

Resources

About