Speech Waveform Synthesis from MFCC Sequences with Generative Adversarial Networks

Keywords

Mel-frequency cepstrum, Computer science, Speech recognition, Cepstrum, Waveform

References

• Introducing CURRENNT: the munich open-source CUDA recurrent neural network toolkitFelix Weninger, Johannes Michael Bergmann, Björn W. Schuller — 2015
• Perceptual Properties of Current Speech Recognition TechnologyHynek Heřmanský, Jordan Cohen, Richard M. Stern — 2013
• Theano: new features and speed improvementsFrédéric Bastien, Pascal Lamblin, Razvan Pascanu — 2012
• Speaker Recognition by Machines and Humans: A tutorial reviewJohn H. L. Hansen, Taufiq Hasan — 2015
• Linear prediction: A tutorial reviewJ. Makhoul — 1975
• Quasi Closed Phase Glottal Inverse Filtering Analysis With Weighted Linear PredictionManu Airaksinen, Tuomo Raitio, Brad H. Story — 2014
• Predicting fundamental frequency from mel-frequency cepstral coefficients to enable speech reconstruction.Xu Shao, Ben Milner — 2005
• Low Bit-Rate Speech Coding Through Quantization of Mel-Frequency Cepstral CoefficientsLaura E. Boucheron, Phillip L. De Leon, Steven Sandoval — 2012
• Non-parametric techniques for pitch-scale and time-scale modification of speechÉric Moulines, Jean Laroche — 1995
• On the inversion of Mel-frequency cepstral coefficients for speech enhancement applicationsLaura E. Boucheron, Phillip L. De Leon — 2008
• An overview of text-independent speaker recognition: From features to supervectorsTomi Kinnunen, Haizhou Li — 2010
• Prediction of Fundamental Frequency and Voicing From Mel-Frequency Cepstral Coefficients for Unconstrained Speech ReconstructionBen Milner, Xu Shao — 2007
• librosa: Audio and Music Signal Analysis in PythonBrian McFee, Colin Raffel, Dawen Liang — 2015
• Mel-generalized cepstral analysis - a unified approach to speech spectral estimationKeiichi Tokuda, Takao Kobayashi, Takashi Masuko — 1994
• High-pitched excitation generation for glottal vocoding in statistical parametric speech synthesis using a deep neural networkLauri Juvela, Bajibabu Bollepalli, Manu Airaksinen — 2016
• Least Squares Generative Adversarial NetworksXudong Mao, Qing Li, Haoran Xie — 2016
• Using Text and Acoustic Features in Predicting Glottal Excitation Waveforms for Parametric Speech Synthesis with Recurrent Neural NetworksLauri Juvela, Xin Wang, Shinji Takaki — 2016
• Non-parallel voice conversion using i-vector PLDA: towards unifying speaker verification and transformationTomi Kinnunen, Lauri Juvela, Paavo Alku — 2017
• Bias and Statistical Significance in Evaluating Speech Synthesis with Mean Opinion ScoresAndrew Rosenberg, Bhuvana Ramabhadran — 2017
• Sequence-to-Sequence Voice Conversion with Similarity Metric Learned Using Generative Adversarial NetworksTakuhiro Kaneko, Hirokazu Kameoka, Kaoru Hiramatsu — 2017

Cited by

• Speaker-independent raw waveform model for glottal excitationLauri Juvela, Vassilis Tsiaras, Bajibabu Bollepalli — 2018
• Waveform Generation for Text-to-speech Synthesis Using Pitch-synchronous Multi-scale Generative Adversarial NetworksLauri Juvela, Bajibabu Bollepalli, Junichi Yamagishi — 2018
• Physiological Waveform Imputation of Missing Data using Convolutional AutoencodersDaniel Miller, Andrew Ward, Nicholas Bambos — 2018
• GELP: GAN-Excited Linear Prediction for Speech Synthesis from Mel-spectrogramLauri Juvela, Bajibabu Bollepalli, Junichi Yamagishi — 2019
• GlotNet—A Raw Waveform Model for the Glottal Excitation in Statistical Parametric Speech SynthesisLauri Juvela, Bajibabu Bollepalli, Vassilis Tsiaras — 2019
• ASVspoof 2019: Future Horizons in Spoofed and Fake Audio DetectionMassimiliano Todisco, Xin Wang, Ville Vestman — 2019
• Constrained Learned Feature Extraction for Acoustic Scene ClassificationTeng Zhang, Ji Wu — 2019
• Vocoder-free text-to-speech synthesis incorporating generative adversarial networks using low-/multi-frequency STFT amplitude spectraYuki Saito, Shinnosuke Takamichi, Hiroshi Saruwatari — 2019
• Analysis by Adversarial Synthesis - A Novel Approach for Speech VocodingAhmed Mustafa, Arijit Biswas, Christian Bergler — 2019
• Deep Residual Neural Networks for Audio Spoofing DetectionMoustafa Alzantot, Ziqi Wang, Mani Srivastava — 2019
• Image-Evoked Affect and its Impact on Eeg-Based BiometricsPablo Arnau‐González, Stamos Katsigiannis, Miguel Arevalillo‐Herráez — 2019
• A Speech Reconstruction Algorithm via Iteratively Reweighted ℓ2 Minimization for MFCC CodecGang Min, Xiongwei Zhang, Xiangyang Liu — 2019
• Neural Source-Filter Waveform Models for Statistical Parametric Speech SynthesisXin Wang, Shinji Takaki, Junichi Yamagishi — 2019
• Blind Vocoder Speech Reconstruction using Generative Adversarial NetworksYoav Blum, David Burshtein — 2019
• Bridging Mixture Density Networks with Meta-Learning for Automatic Speaker IdentificationRuirui Li, Jyun‐Yu Jiang, Xian Wu — 2020
• The Research of Forensic Voiceprint Identification Based on WMFCCGuorui Feng, Xu Chang — 2019
• Adversarially Training for Audio ClassifiersRaymel Alfonso Sallo, Mohammad Esmaeilpour, Patrick Cardinal — 2020
• A retrieval algorithm for encrypted speech based on convolutional neural network and deep hashingQiuyu Zhang, Yuzhou Li, Yingjie Hu — 2020
• Intelligent Instruction-Based IoT Framework for Smart Home Applications using Speech RecognitionYao Ge, Shuja Ansari, Amir M. Abdulghani — 2020
• Environment Sound Classification Based on Visual Multi-Feature Fusion and GRU-AWSNing Peng, Aibin Chen, Guoxiong Zhou — 2020

Related papers

• Pitch synchronous residual excited speech reconstruction on the MFCCPsutka, J. — 2000
• Применение частотного маскирования при MFCC-параметризации речи на фоне шумовK. K. Tomchuk — 2016
• Speech Analysis for Automatic Speech RecognitionNoelia Alcaraz Meseguer — 2009
• Robust Features for Noisy Speech Recognition using MFCC Computation from Magnitude Spectrum of Higher Order Autocorrelation CoefficientsAmita Dev, Poonam Bansal — 2010
• Fundamental frequency and voicing prediction from MFCCs for speech reconstruction from unconstrained speechBen Milner, Xu Shao, Jonathan Darch — 2005
• Predicting fundamental frequency from mel-frequency cepstral coefficients to enable speech reconstruction.Xu Shao, Ben Milner — 2005
• Speech Magnitude Spectrum Reconstruction from MFCCs Using Deep Neural NetworkWenbin Jiang, Peilin Liu, Fei Wen — 2018
• Analysis and prediction of acoustic speech features from mel-frequency cepstral coefficients in distributed speech recognition architectures.Jonathan Darch, Ben Milner, Saeed V. Vaseghi — 2008