International Audio Laboratories Erlangen

A Lightweight Neural TTS system for High-quality German Speech Synthesis

This is the accompanying website for the paper "A lightweight neural TTS system for high-quality speech synthesis in German" by Prachi Govalkar, Ahmed Mustafa, Nicola Pia, Judith Bauer, Metehan Yurt, Yiğitcan Özer and Christian Dittmar at the ITG Conference on Speech Communication 2021.

Abstract

This paper describes a lightweight neural text-to-speech system for the German language. The system is composed of a non-autoregressive spectrogram predictor, followed by a novel vocoder model called StyleMelGAN. Our complete system has a very tiny footprint of 61 MB and is able to synthesize high-quality speech output faster than real-time both on CPU (2.26x) and GPU (44.71x). We additionally propose a modified version of the vocoder called Multi-band StyleMelGAN, which offers a significant improvement in inference speed with a small trade-off in speech quality. In a perceptual listening test with the complete TTS pipeline, the best configuration achieves a mean opinion score of 3.84 using StyleMelGAN, compared to 4.23 for professional speech recordings.

Audio Examples from our Listening Test

Below, we provide all speech items that have been presented in our perceptual listening test to the human raters. The naming of the different conditions is explained in the following table.

Name Description
FT + PGHI ForwardTacotron + Phase Gradient Heap Integration
FT + WGLO ForwardTacotron + WaveGlow
FT + MBSMG ForwardTacotron + Multi-band StyleMelGAN
FT + SMG ForwardTacotron + StyleMelGAN
REF Reference speech recordings

Test Item 001, Female

2021-ITG-TestItem-001-FEMALE-REF

Test Item 001, Male

2021-ITG-TestItem-001-MALE-REF

Test Item 002, Female

2021-ITG-TestItem-002-FEMALE-REF

Test Item 002, Male

2021-ITG-TestItem-002-MALE-REF

Test Item 003, Female

2021-ITG-TestItem-003-FEMALE-REF

Test Item 003, Male

2021-ITG-TestItem-003-MALE-REF

Test Item 004, Female

2021-ITG-TestItem-004-FEMALE-REF

Test Item 004, Male

2021-ITG-TestItem-004-MALE-REF

Test Item 005, Female

2021-ITG-TestItem-005-FEMALE-REF

Test Item 005, Male

2021-ITG-TestItem-005-MALE-REF

Test Item 006, Female

2021-ITG-TestItem-006-FEMALE-REF

Test Item 006, Male

2021-ITG-TestItem-006-MALE-REF

Test Item 007, Female

2021-ITG-TestItem-007-FEMALE-REF

Test Item 007, Male

2021-ITG-TestItem-007-MALE-REF

Test Item 008, Female

2021-ITG-TestItem-008-FEMALE-REF

Test Item 008, Male

2021-ITG-TestItem-008-MALE-REF

Test Item 009, Female

2021-ITG-TestItem-009-FEMALE-REF

Test Item 009, Male

2021-ITG-TestItem-009-MALE-REF

Test Item 010, Female

2021-ITG-TestItem-010-FEMALE-REF

Test Item 010, Male

2021-ITG-TestItem-010-MALE-REF

References

  1. Ryan Prenger, Raffael Valle, and Bryan Catanzaro
    WaveGlow: A Flow-based Generative Network for Speech Synthesis
    In ICASSP: 3617–3621, 2019. DOI 
    @inproceedings{PrengerVC19_WaveGlow_ICASSP,
author={Ryan Prenger and Raffael Valle and Bryan Catanzaro},
booktitle={{ICASSP}},
title={{W}ave{G}low: {A} {F}low-based {G}enerative {N}etwork for {S}peech {S}ynthesis},
year={2019},
address = {Brighton, UK},
month = {May},
pages={3617--3621},
doi  = {10.1109/ICASSP.2019.8683143},
}
  2. Xin Wang, Jaime Lorenzo-Trueba, Shinji Takaki, Lauri Juvela, and Junichi Yamagishi
    A Comparison of Recent Waveform Generation and Acoustic Modeling Methods for Neural-Network-Based Speech Synthesis
    In Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP): 4804–4808, 2018. 
    @inproceedings{WangLTJY18_NeuralVocodersComparison_ICASSP,
author    = {Xin Wang and Jaime Lorenzo{-}Trueba and Shinji Takaki and Lauri Juvela and Junichi Yamagishi},
title     = {{A} {C}omparison of {R}ecent {W}aveform {G}eneration and {A}coustic {M}odeling {M}ethods for {N}eural-{N}etwork-{B}ased {S}peech {S}ynthesis},
booktitle = {Proceedings of the {IEEE} International Conference on Acoustics, Speech, and Signal Processing ({ICASSP})},
address   = {Calgary, Canada},
month     = {April},
pages     = {4804--4808},
year      = {2018},
}
  3. Manu Airaksinen, Lauri Juvela, Bajibabu Bollepalli, Junichi Yamagishi, and Paavo Alku
    A Comparison Between STRAIGHT, Glottal, and Sinusoidal Vocoding in Statistical Parametric Speech Synthesis
    IEEE/ACM Transactions on Audio, Speech, and Language Processing, 26(9): 1658–1670, 2018. 
    @article{AiraksinenJBYA18_VocoderComparison_IEEE-TASLP,
author    = {Manu Airaksinen and Lauri Juvela and Bajibabu Bollepalli and Junichi Yamagishi and Paavo Alku},
title     = {A {C}omparison {B}etween {STRAIGHT}, {G}lottal, and {S}inusoidal {V}ocoding in {S}tatistical {P}arametric {S}peech {S}ynthesis},
journal   = {{IEEE/ACM} Transactions on Audio, Speech, and Language Processing},
volume    = {26},
number    = {9},
pages     = {1658--1670},
year      = {2018},
}
  4. Prachi Govalkar, Johannes Fischer, Frank Zalkow, and Christian Dittmar
    A Comparison of Recent Neural Vocoders for Speech Signal Reconstruction
    In ISCA Speech Synthesis Workshop: 7–12, 2019. DOI 
    @InProceedings{Govalkar2019_NeuralVocoders_SSW,
author    = {Prachi Govalkar and Johannes Fischer and Frank Zalkow and Christian Dittmar},
booktitle = {ISCA Speech Synthesis Workshop},
title     = {{A Comparison of Recent Neural Vocoders for Speech Signal Reconstruction}},
year      = {2019},
address   = {Vienna, Austria},
month     = {September},
pages     = {7--12},
doi       = {10.21437/SSW.2019-2},
}
  5. Daniel W. Griffin and Jae S. Lim
    Signal Estimation from Modified Short-Time Fourier Transform
    IEEE Transactions on Acoustics, Speech, and Signal Processing, 32(2): 236–243, 1984. 
    @article{GriffinL84_SpecgramInversion_TASSP,
author={Daniel W. Griffin and Jae S. Lim},
title={Signal {E}stimation from {M}odified {S}hort-{T}ime {F}ourier {T}ransform},
journal={{IEEE} Transactions on Acoustics, Speech, and Signal Processing},
year={1984},
volume={32},
number={2},
pages={236--243}
}
  6. Aäron van den Oord, Sander Dieleman, Heiga Zen, Karen Simonyan, Oriol Vinyals, Alex Graves, Nal Kalchbrenner, Andrew Senior, and Koray Kavukcuoglu
    WaveNet: A Generative Model for Raw Audio
    In Proceedings of the ISCA Speech Synthesis Workshop: 125–125, 2016. 
    @inproceedings{OordDZSVGKSK16_WaveNet_SSW,
author={A{\"{a}}ron van den Oord and Sander Dieleman and Heiga Zen and Karen Simonyan and Oriol Vinyals and Alex Graves and Nal Kalchbrenner and Andrew Senior and Koray Kavukcuoglu},
title={{W}ave{N}et: {A} {G}enerative {M}odel for {R}aw {A}udio},
year={2016},
booktitle={Proceedings of the {ISCA} Speech Synthesis Workshop},
pages={125--125},
address = {Sunnyvale, CA, USA},
month = {September},
}
  7. Kundan Kumar, Rithesh Kumar, Thibault de Boissiere, Lucas Gestin, Wei Zhen Teoh, Jose Sotelo, Alexandre de Br 'ebisson, Yoshua Bengio, and Aaron C. Courville
    MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesis
    In Proceedings of the Annual Conference on Neural Information Processing Systems (NIPS): 14881–14892, 2019. 
    @inproceedings{KumarKBGTSBBC19_MelGAN_NIPS,
author    = {Kundan Kumar and Rithesh Kumar and Thibault de Boissiere and Lucas Gestin and Wei Zhen Teoh and Jose Sotelo and Alexandre de Br{\'{e}}bisson and Yoshua Bengio and Aaron C. Courville},
title     = {Mel{GAN}: Generative Adversarial Networks for Conditional Waveform Synthesis},
booktitle = {Proceedings of the Annual Conference on Neural Information Processing Systems ({NIPS})},
address = {Vancouver, BC, Canada},
month   =  {December},
pages     = {14881--14892},
year      = {2019},
}
  8. Kingma, Durk P and Dhariwal, Prafulla
    Glow: Generative Flow with Invertible 1x1 Convolutions
    In Proceedings of the Annual Conference on Neural Information Processing Systems (NIPS): 10215–10224, 2018. 
    @inproceedings{KingmaD18_Glow_NIPS,
title={{G}low: {G}enerative {F}low with {I}nvertible 1x1 {C}onvolutions},
author={Kingma, Durk P and Dhariwal, Prafulla},
booktitle={Proceedings of the Annual Conference on Neural Information Processing Systems ({NIPS})},
address = {Montr{\'{e}}al, Canada},
month = {December},
pages={10215--10224},
year={2018}
}
  9. Rec, ITUT
    P. 800: Methods for subjective determination of transmission quality
    International Telecommunication Union, Geneva, 22, 1996. 
    @article{Rec_P800_ITU,
title={P. 800: Methods for subjective determination of transmission quality},
author={Rec, ITUT},
journal={International Telecommunication Union, Geneva},
volume={22},
year={1996}
}
  10. Rec, ITUT
    P. 808: Subjective evaluation of speech quality with a crowdsourcing approach
    International Telecommunication Union, Geneva, 2018. 
    @article{Rec_P808_ITU,
title={P. 808: Subjective evaluation of speech quality with a crowdsourcing approach},
author={Rec, ITUT},
journal={International Telecommunication Union, Geneva},
year={2018}
}
  11. 2005-2018, Amazon Mechanical Turk, Inc.
    https://www.mturk.com/, Accessed: 2020-09-15, . 
    @misc{MTurk,
title = {2005-2018, {A}mazon {M}echanical {T}urk, {I}nc.},
howpublished =  {\url{https://www.mturk.com/}},
note = {Accessed: 2020-09-15}
}
  12. 2020, Amazon Web Services, Inc.
    https://aws.amazon.com/, Accessed: 2020-08-19, . 
    @misc{Aws,
title = {2020, {A}mazon {W}eb {S}ervices, {I}nc.},
howpublished =  {\url{https://aws.amazon.com/}},
note = {Accessed: 2020-08-19}
}
  13. Michael Schoeffler, Fabian-Robert Stöter, Bernd Edler, and Jürgen Herre
    Towards the Next Generation of Web-based Experiments: A Case Study Assessing Basic Audio Quality Following the ITU-R Recommendation BS.1534 (MUSHRA)
    In Web Audio Conference, 2015. 
    @conference{SchoefflerSEH15_WebMUSHRA_WAC,
author = {Michael Schoeffler and Fabian-Robert St\"{o}ter and Bernd Edler and J\"{u}rgen Herre},
booktitle = {Web Audio Conference},
address = {Paris, France},
month = {January},
title = {{{T}owards the {N}ext {G}eneration of {W}eb-based {E}xperiments: {A} {C}ase {S}tudy {A}ssessing {B}asic {A}udio {Q}uality {F}ollowing the {ITU}-{R} {R}ecommendation {BS}.1534 ({MUSHRA})}},
year = {2015},
}
  14. Jonathan Shen, Ruoming Pang, Ron J. Weiss, Mike Schuster, Navdeep Jaitly, Zongheng Yang, Zhifeng Chen, Yu Zhang, Yuxuan Wang, R. J. Skerry-Ryan, Rif A. Saurous, Yannis Agiomyrgiannakis, and Yonghui Wu
    Natural TTS Synthesis by Conditioning Wavenet on Mel-Spectrogram Predictions
    In ICASSP: 4779–4783, 2018. 
    @inproceedings{Shen2018_Natural_ICASSP,
author    = {Jonathan Shen and Ruoming Pang and Ron J. Weiss and Mike Schuster and Navdeep Jaitly and Zongheng Yang and Zhifeng Chen and Yu Zhang and Yuxuan Wang and R. J. Skerry{-}Ryan and Rif A. Saurous and Yannis Agiomyrgiannakis and Yonghui Wu},
title     = {{Natural {TTS} Synthesis by Conditioning Wavenet on {Mel}-Spectrogram Predictions}},
booktitle = {{ICASSP}},
address   = {Calgary, AB, Canada},
month     = {April},
pages     = {4779--4783},
year      = {2018},
}
  15. Zden vek Pru vsa, P 'eter Balázs, and Peter L. S ondergaard
    A Noniterative Method for Reconstruction of Phase From STFT Magnitude
    IEEE/ACM Transactions on Audio, Speech, and Language Processing, 25(5): 1154–1164, 2017. 
    @article{Prusa2017_PGHI_IEEE,
author    = {Zden\v{e}k Pru\v{s}a and P{\'{e}}ter Bal{\'{a}}zs and Peter L. S{\o}ndergaard},
title     = {{A} {N}oniterative {M}ethod for {R}econstruction of {P}hase {F}rom {STFT} {M}agnitude},
journal   = {{IEEE/ACM} Transactions on Audio, Speech, and Language Processing},
volume    = {25},
number    = {5},
pages     = {1154--1164},
year      = {2017},
}
  16. Zeng, Zhen, Wang, Jianzong, Cheng, Ning, Xia, Tian, and Xiao, Jing
    Aligntts: Efficient Feed-Forward Text-to-Speech System Without Explicit Alignment
    In ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP): 6714–6718, 2020. DOI 
    @inproceedings{zeng2020aligntts,
author={Zeng, Zhen and Wang, Jianzong and Cheng, Ning and Xia, Tian and Xiao, Jing},
booktitle={ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
title={Aligntts: Efficient Feed-Forward Text-to-Speech System Without Explicit Alignment},
year={2020},
volume={},
number={},
pages={6714-6718},
doi={10.1109/ICASSP40776.2020.9054119}
}
  17. Jeff Donahue, Sander Dieleman, Mikolaj Binkowski, Erich Elsen, and Karen Simonyan
    End-to-End Adversarial Text-to-Speech
    CoRR, abs/2006.03575, 2020. 
    @article{donahue2020end,
author    = {Jeff Donahue and Sander Dieleman and Mikolaj Binkowski and Erich Elsen and Karen Simonyan},
title     = {End-to-End Adversarial Text-to-Speech},
journal   = {CoRR},
volume    = {abs/2006.03575},
year      = {2020},
url       = {https://arxiv.org/abs/2006.03575},
archivePrefix = {arXiv},
eprint    = {2006.03575},
timestamp = {Tue, 09 Jun 2020 16:38:02 +0200},
biburl    = {https://dblp.org/rec/journals/corr/abs-2006-03575.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
  18. Chenfeng Miao, Shuang Liang, Zhencheng Liu, Minchuan Chen, Jun Ma, Shaojun Wang, and Jing Xiao
    EfficientTTS: An Efficient and High-Quality Text-to-Speech Architecture
    CoRR, 2012.03500, 2020. 
    @article{miao2020efficienttts,
title={EfficientTTS: An Efficient and High-Quality Text-to-Speech Architecture},
author={Chenfeng Miao and Shuang Liang and Zhencheng Liu and Minchuan Chen and Jun Ma and Shaojun Wang and Jing Xiao},
year={2020},
journal   = {CoRR},
volume    = {2012.03500},
archivePrefix={arXiv},
primaryClass={eess.AS}
}
  19. Dan Lim, Won Jang, Gyeonghwan O, Heayoung Park, Bongwan Kim, and Jaesam Yoon
    JDI-T: Jointly trained Duration Informed Transformer for Text-To-Speech without Explicit Alignment
    CoRR, 2005.07799, 2020. 
    @article{lim2020jdi,
title = {JDI-T: Jointly trained Duration Informed Transformer for Text-To-Speech without Explicit Alignment},
author = {Dan Lim and Won Jang and Gyeonghwan O and Heayoung Park and Bongwan Kim and Jaesam Yoon},
journal   = {CoRR},
year = {2020},
volume = {2005.07799}
}
  20. Yi Ren, Yangjun Ruan, Xu Tan, Tao Qin, Sheng Zhao, Zhou Zhao, and Tie-Yan Liu
    FastSpeech: Fast, Robust and Controllable Text to Speech
    CoRR, abs/1905.09263, 2019. 
    @article{ren2019fastspeech,
author    = {Yi Ren and Yangjun Ruan and Xu Tan and Tao Qin and Sheng Zhao and Zhou Zhao and Tie{-}Yan Liu},
title     = {FastSpeech: Fast, Robust and Controllable Text to Speech},
journal   = {CoRR},
volume    = {abs/1905.09263},
year      = {2019},
url       = {http://arxiv.org/abs/1905.09263},
archivePrefix = {arXiv},
eprint    = {1905.09263},
timestamp = {Wed, 11 Nov 2020 08:48:07 +0100},
biburl    = {https://dblp.org/rec/journals/corr/abs-1905-09263.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
  21. Park, Taesung, Liu, Ming-Yu, Wang, Ting-Chun, and Zhu, Jun-Yan
    Semantic image synthesis with spatially-adaptive normalization
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR): 2337–2346, 2019. 
    @inproceedings{park2019semantic,
title={Semantic image synthesis with spatially-adaptive normalization},
author={Park, Taesung and Liu, Ming-Yu and Wang, Ting-Chun and Zhu, Jun-Yan},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
pages={2337--2346},
year={2019}
}
  22. Mikolaj Binkowski, Jeff Donahue, Sander Dieleman, Aidan Clark, Erich Elsen, Norman Casagrande, Luis C. Cobo, and Karen Simonyan
    High Fidelity Speech Synthesis with Adversarial Networks
    CoRR, abs/1909.11646, 2019. 
    @article{binkowski2019high,
author    = {Mikolaj Binkowski and Jeff Donahue and Sander Dieleman and Aidan Clark and Erich Elsen and Norman Casagrande and Luis C. Cobo and Karen Simonyan},
title     = {High Fidelity Speech Synthesis with Adversarial Networks},
journal   = {CoRR},
volume    = {abs/1909.11646},
year      = {2019},
url       = {http://arxiv.org/abs/1909.11646},
archivePrefix = {arXiv},
eprint    = {1909.11646},
timestamp = {Fri, 27 Sep 2019 13:04:21 +0200},
biburl    = {https://dblp.org/rec/journals/corr/abs-1909-11646.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
  23. Cuturi, Marco and Blondel, Mathieu
    Soft-dtw: a differentiable loss function for time-series
    In International Conference on Machine Learning: 894–903, 2017. 
    @inproceedings{cuturi2017soft,
title={Soft-dtw: a differentiable loss function for time-series},
author={Cuturi, Marco and Blondel, Mathieu},
booktitle={International Conference on Machine Learning},
pages={894--903},
year={2017},
organization={PMLR}
}
  24. Yuxuan Wang, R.J. Skerry-Ryan, Daisy Stanton, Yonghui Wu, Ron J. Weiss, Navdeep Jaitly, Zongheng Yang, Ying Xiao, Zhifeng Chen, Samy Bengio, Quoc Le, Yannis Agiomyrgiannakis, Rob Clark, and Rif A. Saurous
    Tacotron: Towards End-to-End Speech Synthesis
    In Proc. Interspeech 2017: 4006–4010, 2017. DOI 
    @inproceedings{wang2017tacotron,
author = {Yuxuan Wang and R.J. Skerry-Ryan and Daisy Stanton and Yonghui Wu and Ron J. Weiss and Navdeep Jaitly and Zongheng Yang and Ying Xiao and Zhifeng Chen and Samy Bengio and Quoc Le and Yannis Agiomyrgiannakis and Rob Clark and Rif A. Saurous},
title = {Tacotron: Towards End-to-End Speech Synthesis},
year = 2017,
booktitle = {Proc. Interspeech 2017},
pages = {4006--4010},
doi = {10.21437/Interspeech.2017-1452},
url = {http://dx.doi.org/10.21437/Interspeech.2017-1452}
}
  25. Schaefer, C.
    ForwardTacotron
    https://github.com/as-ideas/ForwardTacotron, Accessed: 2020-06, 2020. 
    @misc{Schaefer2020,
author = {Schaefer, C.},
title = {ForwardTacotron},
year = {2020},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/as-ideas/ForwardTacotron}},
note = {Accessed: 2020-06}
}
  26. Dmitry Ulyanov, Andrea Vedaldi, and Victor S. Lempitsky
    Instance Normalization: The Missing Ingredient for Fast Stylization
    CoRR, abs/1607.08022, 2016. 
    @article{ulyanov2016instance,
author    = {Dmitry Ulyanov and Andrea Vedaldi and Victor S. Lempitsky},
title     = {Instance Normalization: The Missing Ingredient for Fast Stylization},
journal   = {CoRR},
volume    = {abs/1607.08022},
year      = {2016},
url       = {http://arxiv.org/abs/1607.08022},
archivePrefix = {arXiv},
eprint    = {1607.08022},
timestamp = {Mon, 13 Aug 2018 16:47:58 +0200},
biburl    = {https://dblp.org/rec/journals/corr/UlyanovVL16.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
}
  27. Mustafa, Ahmed, Pia, Nicola, and Fuchs, Guillaume
    StyleMelGAN: An Efficient High-Fidelity Adversarial Vocoder with Temporal Adaptive Normalization
    In ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP): 6034–6038, 2021. DOI 
    @inproceedings{mustafa2020stylemelgan,
author={Mustafa, Ahmed and Pia, Nicola and Fuchs, Guillaume},
booktitle={ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
title={StyleMelGAN: An Efficient High-Fidelity Adversarial Vocoder with Temporal Adaptive Normalization},
year={2021},
volume={},
number={},
pages={6034-6038},
doi={10.1109/ICASSP39728.2021.9413605},
}
  28. Yang, Geng, Yang, Shan, Liu, Kai, Fang, Peng, Chen, Wei, and Xie, Lei
    Multi-band MelGAN: Faster waveform generation for high-quality text-to-speech
    In 2021 IEEE Spoken Language Technology Workshop (SLT): 492–498, 2021. 
    @inproceedings{yang2021multi,
title={Multi-band MelGAN: Faster waveform generation for high-quality text-to-speech},
author={Yang, Geng and Yang, Shan and Liu, Kai and Fang, Peng and Chen, Wei and Xie, Lei},
booktitle={2021 IEEE Spoken Language Technology Workshop (SLT)},
pages={492--498},
year={2021},
organization={IEEE}
}
  29. Yamamoto, Ryuichi, Song, Eunwoo, and Kim, Jae-Min
    Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogram
    In ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP): 6199–6203, 2020. 
    @inproceedings{yamamoto2020parallel,
title={Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogram},
author={Yamamoto, Ryuichi and Song, Eunwoo and Kim, Jae-Min},
booktitle={ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
pages={6199--6203},
year={2020},
organization={IEEE}
}
  30. Nguyen, Truong Q
    Near-perfect-reconstruction pseudo-QMF banks
    IEEE Transactions on signal processing, 42(1): 65–76, 1994. 
    @article{nguyen1994near,
title={Near-perfect-reconstruction pseudo-QMF banks},
author={Nguyen, Truong Q},
journal={IEEE Transactions on signal processing},
volume={42},
number={1},
pages={65--76},
year={1994},
publisher={IEEE}
}
  31. A. Gritsenko, Tim Salimans, Rianne van den Berg, Jasper Snoek, and Nal Kalchbrenner
    A Spectral Energy Distance for Parallel Speech Synthesis
    CoRR, abs/2008.01160, 2020. 
    @article{gritsenko2020spectral,
title={A Spectral Energy Distance for Parallel Speech Synthesis},
author={A. Gritsenko and Tim Salimans and Rianne van den Berg and Jasper Snoek and Nal Kalchbrenner},
journal={CoRR},
year={2020},
volume={abs/2008.01160}
}