ExcitNet Vocoder: A Neural Excitation Model for Parametric Speech Synthesis Systems

By Eunwoo Song, Kyungguen Byun, Hong-Goo Kang. 2018

30 citations recorded by the source corpus.

This paper proposes a WaveNet-based neural excitation model (ExcitNet) for statistical parametric speech synthesis systems. Conventional WaveNet-based neural vocoding systems significantly improve the perceptual quality of synthesized speech by statistically generating a time sequence of speech waveforms through an auto-regressive framework. However, they often suffer from noisy outputs because of the difficulties in capturing the complicated time-varying nature of speech signals. To improve modeling efficiency, the proposed ExcitNet vocoder employs an adaptive inverse filter to decouple spectral components from the speech signal. The residual component, i.e. excitation signal, is then trained and generated within the WaveNet framework. In this way, the quality of the synthesized speech signal can be further improved since the spectral component is well represented by a deep learning framework and, moreover, the residual component is efficiently generated by the WaveNet framework. Experimental results show that the proposed ExcitNet vocoder, trained both speaker-dependently and speaker-independently, outperforms traditional linear prediction vocoders and similarly configured conventional WaveNet vocoders.

Subfields: Artificial Intelligence, Natural Language Processing, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Signal Processing

Citations in the corpus

WaveGrad: Estimating Gradients for Waveform Generation (2020, methodology): “WaveG-rad estimates the gradients of the data density, leveraging on the diffusion probabilistic model (Ho et al., 2020) and score matching framework (Song et al., 2019b; Song & Ermon, 2020).”
Parallel Wavegan: A Fast Waveform Generation Model Based on Generative Adversarial Networks with Multi-Resolution Spectrogram (2019, methodology): “Remarkably, autoregressive generative models such as WaveNet have shown much superior performance over traditional parametric vocoders [4–8].”
Probability density distillation with generative adversarial networks for high-quality parallel waveform generation (2019, methodology): “By appropriately conditioning the acoustic features to the input, WaveNet has also been successfully adopted in a neural vocoder structure for statistical parametric speech synthesis systems [2–5], and end-to-end speech synthesis systems [6–10].”
Neural Text-to-Speech with a Modeling-by-Generation Excitation Vocoder (2020, methodology): “The MbG structure can be extended to any neural excitation vocoder that uses LP coefﬁcients, but the focus here is on the WaveNet-based Ex-citNet vocoder [7].”
Improving LPCNET-Based Text-to-Speech with Linear Prediction-Structured Mixture Density Network (2020, methodology): “ucture to generate the glottal exci1 The term prediction is deﬁned as an intermediate LP operation [5], which will be discussed in detail in section 3. arXiv:2001.11686v1 [eess.AS] 31 Jan 2020 tation [9,10]. In case of the LPCNet, it employed the lightweight WaveRNN model for fast generation of the ”
Lightweight LPCNet-Based Neural Vocoder with Tensor Decomposition (2020, methodology): “In addition, combining the classical source-ﬁlter model with the neural vocoder provides better results, even with a small number of parameters [10, 11, 12].”
Gaussian Lpcnet for Multisample Speech Synthesis (2020, unknown): “However, in many recent studies very good results have been achieved by combining deep learning approach with classical Source-Filter model of speech production (e.g. [5, 12, 13, 14]).”
FeatherWave: An efficient high-fidelity neural vocoder with multi-band linear prediction (2020, methodology): “Similar to GlotNet [14] and ExcitNet [15] which use the WaveNet to capture the glottal excitation signal, the LPCNet models the discretized excitation signal of LPC ﬁl-ters with a WaveRNN for efﬁcient generation.”