GlotNet—A Raw Waveform Model for the Glottal Excitation in Statistical Parametric Speech Synthesis

By Lauri Juvela, Bajibabu Bollepalli, Vassilis Tsiaras, Paavo Alku. 2019

44 citations recorded by the source corpus.

Recently, generative neural network models which operate directly on raw audio, such as WaveNet, have improved the state of the art in text-to-speech synthesis (TTS). Moreover, there is increasing interest in using these models as statistical vocoders for generating speech waveforms from various acoustic features. However, there is also a need to reduce the model complexity, without compromising the synthesis quality. Previously, glottal pulseforms (i.e., time-domain waveforms corresponding to the source of human voice production mechanism) have been successfully synthesized in TTS by glottal vocoders using straightforward deep feedforward neural networks. Therefore, it is natural to extend the glottal waveform modeling domain to use the more powerful WaveNet-like architecture. Furthermore, due to their inherent simplicity, glottal excitation waveforms permit scaling down the waveform generator architecture. In this study, we present a raw waveform glottal excitation model, called GlotNet, and compare its performance with the corresponding direct speech waveform model, WaveNet, using equivalent architectures. The models are evaluated as part of a statistical parametric TTS system. Listening test results show that both approaches are rated highly in voice similarity to the target speaker, and obtain similar quality ratings with large models. Furthermore, when the model size is reduced, the quality degradation is less severe for GlotNet.

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

1. Xu Tan
2. Tao Qin
3. Tie-Yan Liu
4. Frank K. Soong
5. Lei He
6. Junichi Yamagishi
7. Yu Zhang
8. Sheng Zhao

In Natural Language Processing

1. Tao Qin
2. Tie-Yan Liu
3. Xu Tan
4. Lei He
5. Junichi Yamagishi
6. Frank K. Soong
7. Yu Zhang
8. Heiga Zen

In Signal Processing

1. Xu Tan
2. Tao Qin
3. Frank K. Soong
4. Tie-Yan Liu
5. Junichi Yamagishi
6. Lei He
7. Yu Zhang
8. Heiga Zen

Citations in the corpus

• WaveGrad: Estimating Gradients for Waveform Generation (2020, background): “, 2020), and the speech production mechanism (Juvela et al., 2019; Wang et al., 2020).”
• A Survey on Neural Speech Synthesis (2021, background): “Other Vocoders Some works leverage neural-based source-filter model for waveform generation [375, 374, 371, 207, 146, 145, 75, 305, 407], aiming to achieve high voice quality while maintaining controllable speech generation.”
• Neural Analysis and Synthesis: Reconstructing Speech from Self-Supervised Representations (2021, methodology): “By modeling source and filter components with deep networks, it has been used for applications such as vocoder [49, 23, 45] and acoustic feature generation [25].”
• An Efficient Subband Linear Prediction for LPCNet-Based Neural Synthesis (2020, methodology): “Based on the source-filter model, some neural vocoders can get nat-ural speech with smaller model and higher efficiency, such as LPCNet[8], LP-WaveNet[9], neural source-filter model[10], and glottal neural vocoders[11, 12, 13, 14].”
• GELP: GAN-Excited Linear Prediction for Speech Synthesis from Mel-spectrogram (2019, methodology): “Meanwhile, our version of WaveNet has previously achieved MOS scores above 4 using glottal vocoder acoustic features [17].”
• GELP: GAN-Excited Liner Prediction for Speech Synthesis from Mel-spectrogram (2019, methodology): “Mean-while, the source-filter model has also been used with linear predictive (LP) filters to build neural models for the glottal excitation [16, 17] or the LP residual signal [18, 19].”
• Improving LPCNET-Based Text-to-Speech with Linear Prediction-Structured Mixture Density Network (2020, unknown): “tation [9, 10].”
• A Neural Vocoder With Hierarchical Generation of Amplitude and Phase Spectra for Statistical Parametric Speech Synthesis (2019, methodology): “as glottal neural vocoder [26]–[28], LP-WaveNet [29], LPCNet [30], and neural source-filter (NSF) vocoder [31], have been further proposed by combining speech production mechanisms”
• 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.”