Voice Conversion Using Deep Neural Networks With Layer-Wise Generative Training

By Linghui Chen, Zhenhua Ling, Li-Juan Liu, Lirong Dai. 2014

235 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

Neural Voice Cloning with a Few Samples (2018, methodology): “There are also approaches to model spectral conversion using neural networks (Desai et al., 2010; Chen et al., 2014; Hwang et al., 2015).”
Adaptive Wavenet Vocoder for Residual Compensation in GAN-Based Voice Conversion (2018, background): “Recently, deep learning approaches such as restricted Boltzmann machine (RBM) [12, 13], long short-term memory (LSTM) [14, 15], and autoencoder [16, 17] are shown to be effective in modeling the relationship between source and target features more accurately than conventional techniques.”
StarGAN-VC: non-parallel many-to-many Voice Conversion Using Star Generative Adversarial Networks (2018, background): “Recently, some attempts have been made to develop non-parallel meth-ods [17–29].”
NoiseVC: Towards High Quality Zero-Shot Voice Conversion (2021, background): “Most of previous works mainly focus on supervised VC [1, 2, 3], and achieve satisfactory performance.”
Locally Linear Embedding for Exemplar-Based Spectral Conversion (2016, methodology): “Numerous VC approaches have been proposed, such as Gaussian mixture model (GMM)-based [1-4], frequency warpingbased [5, 6], neural networks (NN)-based [7-10], and exemplar-based [11-13] approaches.”
Voice Impersonation Using Generative Adversarial Networks (2018, methodology): “The strategies used range from simple codebookbased conversion [4] and minimum-mean-squared error linear estimators [5] to sophisticated neural network models [6].”
A Regression Approach to Speech Enhancement Based on Deep Neural Networks (2015, methodology): “Although these traditional MMSE-based methods are able to yield lower musical noise (e.g., [10], [11]), a trade-off in reducing speech distortion and residual noise needs to be made due to the sophisticated statistical properties of the interactions between speech and noise signals.”
A Deep Generative Architecture for Postfiltering in Statistical Parametric Speech Synthesis (2015, methodology): “However, previous work in VC has indicated that listeners prefer synthetic speech generated using a network architecture without the ﬁne-tuning over one using the ﬁne-tuning based on the MMSE criterion [25].”
Multi-target Voice Conversion without Parallel Data by Adversarially Learning Disentangled Audio Representations (2018, background): “In general, among the difﬁcult problems for VC approaches, the need of aligned data [11, 12, 13, 14, 15, 16, 17], and over-smoothing of signals [1, 18, 19] are two examples carefully studied.”
A probabilistic interpretation for artificial neural network-based voice conversion (2015, methodology): “Moreover, the advantage of using the MLPG algorithm for NN-based speech parameter generation has already been reported in previous studies ([7-9] and [15, 16]); similar trends are also observed in our preliminary experiments.”
WaveNet Vocoder with Limited Training Data for Voice Conversion (2018, methodology): “Various deep neural networks (DNN) were designed to boost the ability of conventional GMMs [6, 7, 8, 9, 10].”
An Overview of Voice Conversion and Its Challenges: From Statistical Modeling to Deep Learning (2020, background): “[171] proposed to model the distributions of spectral envelopes of source and target speakers respectively through a layer-wise”
CycleGAN-VC: Non-parallel Voice Conversion Using Cycle-Consistent Adversarial Networks (2018, methodology): “Neural network (NN)-based methods, such as a restricted Boltzmann machine (RBM) [12], [13], feed forward NN [14]–[16], recurrent NN (RNN) [17], [18], convolutional NN (CNN) [8], and generative adversarial network (GAN) [8], and exemplar-based methods, such as non-negative matrix factorization (NMF) ”
Modeling and Transforming Speech Using Variational Autoencoders (2016, methodology): “These models have been used for tasks such as speech recognition [7], speech synthesis [8], and other tasks like voice conversion [9].”
Joint training framework for text-to-speech and voice conversion using multi-source Tacotron and WaveNet (2019, background): “Dynamic kernel partial least squares regression (DKPLS) [7] and neural network based approaches [8, 9, 10, 11, 12] learn non-linear transforms.”