An investigation of multi-speaker training for wavenet vocoder

By Tomoki Hayashi, Akira Tamamori, Kazuhiro Kobayashi, Kazuya Takeda, Tomoki Toda. 2017

100 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Signal Processing

Citations in the corpus

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): “ter size was 3. The conditioning features were upsampled by nearest neighbor upsampling followed by 2-D convolutions for interpolation [32]. The upsampling was split into ﬁve modules. The scales were [2, 2, 2, 3, 5]. The kernel sizes for the 2-D convolutions were set to 2 s+ 1, where denotes the ups”
ExcitNet Vocoder: A Neural Excitation Model for Parametric Speech Synthesis Systems (2018, methodology): “First, the SI-ExcitNet performed similarly to the ITFTE vocoder but performed much better than the WORLD system in analysis/synthesis.”
Pretraining Techniques for Sequence-to-Sequence Voice Conversion (2020, background): “2 Since speaker-dependent neural vocoders outperform speaker-independent ones [77], we”
Fre-GAN: Adversarial Frequency-consistent Audio Synthesis (2021, methodology): “As an objective quality evaluation, we used MCD13 [34], RMSEf0 [35], and FDSD [36].”
An Investigation of Subband Wavenet Vocoder Covering Entire Audible Frequency Range with Limited Acoustic Features (2018, methodology): “By introducing a noise shaping method [20], it outperformed the conventional STRAIGHT vocoder [3] with a sampling frequency of 16 kHz [21].”
Modeling Prosodic Phrasing With Multi-Task Learning in Tacotron-Based TTS (2020, background): “The recent progress in neural vocoder [4], [11]–[15] also contributes to the improvement of speech quality.”
Knowledge-and-Data-Driven Amplitude Spectrum Prediction for Hierarchical Neural Vocoders (2020, methodology): “Neural vocoders based on these waveform generation models [11, 12, 13, 14, 15, 16] have been developed to reconstruct speech waveforms from various acoustic features for SPSS and some other tasks, such”
Adaptive Wavenet Vocoder for Residual Compensation in GAN-Based Voice Conversion (2018, methodology): “independent WaveNet vocoders [27, 29] were studied and shown initial success.”
Wasserstein GAN and Waveform Loss-Based Acoustic Model Training for Multi-Speaker Text-to-Speech Synthesis Systems Using a WaveNet Vocoder (2018, methodology): “ multi-speaker modeling with speech data in different languages. Multi-speaker speech synthesis has also been investigated in the recent WaveNet-based approaches and in end-to-end approaches. Hayashi [28] attempted WaveNet vocoder-based multi-speaker synthesis using four speakers from the CMU arctic”
An Overview of Voice Conversion and Its Challenges: From Statistical Modeling to Deep Learning (2020, methodology): “WaveNet vocoder [54] outperforms remarkably the traditional parametric vocoders in terms of sound quality.”
Universal Neural Vocoding with Parallel Wavenet (2021, background): “There are also reports of neural vocoders that are capable of synthesising unseen speakers or styles without having to explicitly model speaker identities [17, 10, 9, 11, 18].”
Speaker-independent raw waveform model for glottal excitation (2018, methodology): “speaker-independent models have used the softmax training approach [7, 6].”
Speaker Conditional WaveRNN: Towards Universal Neural Vocoder for Unseen Speaker and Recording Conditions (2020, background): “Previous studies also attempted to improve adaptation capabilities of vocoders [24], either with or without providing speaker information [25, 26].”
Quasi-Periodic Parallel WaveGAN Vocoder: A Non-autoregressive Pitch-dependent Dilated Convolution Model for Parametric Speech Generation (2020, methodology): “The NN-based vocoders [8–11] are also proposed on the basis of these AR SS models to replace the synthesizers of the conventional vocoders for recovering the lost phase information and temporal details and generating highquality speech.”
Training Multi-Speaker Neural Text-to-Speech Systems using Speaker-Imbalanced Speech Corpora (2019, methodology): “A speaker-independent WaveNet vocoder [26] was trained using the combined training data of all speakers.”
WaveNet Vocoder with Limited Training Data for Voice Conversion (2018, methodology): “The method of building speaker-independent WaveNet vocoders was then studied [22].”