Fftnet: A Real-Time Speaker-Dependent Neural Vocoder

By Zeyu Jin, Adam Finkelstein, Gautham J. Mysore, Jingwan Lu. 2018

107 citations recorded by the source corpus.

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. Junichi Yamagishi
5. Frank K. Soong
6. Wei-Ning Hsu
7. Yossi Adi
8. Hung-yi Lee

In Natural Language Processing

1. Junichi Yamagishi
2. Zhenhua Ling
3. Tao Qin
4. Tie-Yan Liu
5. Frank K. Soong
6. Xu Tan
7. Tomoki Toda
8. Xin Eric Wang

In Signal Processing

1. Junichi Yamagishi
2. Xu Tan
3. Frank K. Soong
4. Tao Qin
5. Tie-Yan Liu
6. Yossi Adi
7. Wei-Ning Hsu
8. Zhenhua Ling

Citations in the corpus

• Waveglow: A Flow-based Generative Network for Speech Synthesis (2018, background): “Since then, several auto-regressive models have attempted to speed up inference while retaining quality [6, 20, 21].”
• SoundStream: An End-to-End Neural Audio Codec (2021, methodology): “Other approaches involve lightweight and sparse models [13] and networks mimicking the fast Fourier transform as part of the model [7], [14].”
• A Survey on Neural Speech Synthesis (2021, methodology): “FFTNet [142] uses a simple architecture to mimic the Fast Fourier Transform (FFT), which can generate audio samples in real-time.”
• Speech Resynthesis from Discrete Disentangled Self-Supervised Representations (2021, methodology): “In [18] the authors proposed a VQ-VAE model followed by an FFTNet vocoder model [19].”
• On Generative Spoken Language Modeling from Raw Audio (2021, positive): “(2019) suggested a VQ-VAE model followed by a FFTNet vocoder model (Jin et al., 2018).”
• LPCNET: Improving Neural Speech Synthesis through Linear Prediction (2018, methodology): “One possible remedy – which we did not yet investigate – consists of using post-denoising techniques, as suggested in [6].”
• A Real-Time Wideband Neural Vocoder at 1.6 kb/s Using LPCNet (2019, methodology): “To make the network more robust to the mismatch, we add noise to the training data, as originally suggested in [19].”
• Neural Source-Filter Waveform Models for Statistical Parametric Speech Synthesis (2019, methodology): “Other models, such as WaveRNN [25], FFTNet [26], and subband WaveNet [27], use different network architectures to reduce or parallelize the computation load, but their generation time is still linearly proportional to the waveform length.”
• Neural Feature Predictor and Discriminative Residual Coding for Low-Bitrate Speech Coding (2022, background): “Therefore, we add noise to the input during training for a more robust development, as suggested in [7, 28].”
• Towards Robust Neural Vocoding for Speech Generation: A Survey (2019, methodology): “Neural vocoders can successfully model the data distribution of human voice with acoustic features [13, 14, 15, 16, 17], however the generated speech quality is still restricted by the consistency of training and testing domain due to deep learning’s data-driven”
• Tacotron-Based Acoustic Model Using Phoneme Alignment for Practical Neural Text-to-Speech Systems (2019, methodology): “The first is AR models with simple structures, such as FFTNet [43, 44], WaveRNN [45], and LPCNet [46].”
• Speaker-independent raw waveform model for glottal excitation (2018, result): “Maximum posterior sampling in voiced regions has been re-ported to improve perceived synthetic speech quality [5, 23], and we observed a similar effect in our informal experiments.”
• Improving FFTNet Vocoder with Noise Shaping and Subband Approaches (2018, methodology): “An alternative raw audio autoregressive generative model, FFT-Net [42], has a simpler structure based on a 1 × 1 convolutional network and rectified liner unit (ReLU) layers.”
• Review of end-to-end speech synthesis technology based on deep learning (2021, methodology): “Jin Z, Finkelstein A, Mysore GJ, Lu J (2018) Fftnet: A real-time speaker-dependent neural vocoder.”