An effective approach for voice conversion (VC) is to disentangle linguistic content from other components in the speech signal. The effectiveness of variational autoencoder (VAE) based VC (VAE-VC), for instance, strongly relies on this principle. In our prior work, we proposed a cross-domain VAE-VC (CDVAE-VC) framework, which utilized acoustic features of different properties, to improve the performance of VAE-VC. We believed that the success came from more disentangled latent representations. In this paper, we extend the CDVAE-VC framework by incorporating the concept of adversarial learning, in order to further increase the degree of disentanglement, thereby improving the quality and similarity of converted speech. More specifically, we first investigate the effectiveness of incorporating the generative adversarial networks (GANs) with CDVAE-VC. Then, we consider the concept of domain adversarial training and add an explicit constraint to the latent representation, realized by a speaker classifier, to explicitly eliminate the speaker information that resides in the latent code. Experimental results confirm that the degree of disentanglement of the learned latent representation can be enhanced by both GANs and the speaker classifier. Meanwhile, subjective evaluation results in terms of quality and similarity scores demonstrate the effectiveness of our proposed methods.
Vinai Research（维奈研究院）；Deezer Research and Development（迪泽研发中心）
Voice conversion is a challenging task which transforms the voice characteristics of a source speaker to a target speaker without changing linguistic content. Recently, there have been many works on many-to-many Voice Conversion (VC) based on Variational Autoencoder (VAEs) achieving good results, however, these methods lack the ability to disentangle speaker identity and linguistic content to achieve good performance on unseen speaker scenarios. In this paper, we propose a new method based on feature disentanglement to tackle many to many voice conversion. The method has the capability to dise…
Recently, the standard variational autoencoder has been successfully used to learn a probabilistic prior over speech signals, which is then used to perform speech enhancement. Variational autoencoders have then been conditioned on a label describing a high-level speech attribute (e.g. speech activity) that allows for a more explicit control of speech generation. However, the label is not guaranteed to be disentangled from the other latent variables, which results in limited performance improvements compared to the standard variational autoencoder. In this work, we propose to use an adversarial…
Singapore University of Technology and Design（新加坡科技设计大学）；National University of Singapore（新加坡国立大学）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
Expressive voice conversion performs identity conversion for emotional speakers by jointly converting speaker identity and emotional style. Due to the hierarchical structure of speech emotion, it is challenging to disentangle the emotional style for different speakers. Inspired by the recent success of speaker disentanglement with variational autoencoder (VAE), we propose an any-to-any expressive voice conversion framework, that is called StyleVC. StyleVC is designed to disentangle linguistic content, speaker identity, pitch, and emotional style information. We study the use of style encoder t…
Voice conversion (VC) is a task that transforms voice from target audio to source without losing linguistic contents, it is challenging especially when source and target speakers are unseen during training (zero-shot VC). Previous approaches require a pre-trained model or linguistic data to do the zero-shot conversion. Meanwhile, VC models with Vector Quantization (VQ) or Instance Normalization (IN) are able to disentangle contents from audios and achieve successful conversions. However, disentanglement in these models highly relies on heavily constrained bottleneck layers, thus, the sound qua…
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）
SpeechSplit can perform aspect-specific voice conversion by disentangling speech into content, rhythm, pitch, and timbre using multiple autoencoders in an unsupervised manner. However, SpeechSplit requires careful tuning of the autoencoder bottlenecks, which can be time-consuming and less robust. This paper proposes SpeechSplit 2.0, which constrains the information flow of the speech component to be disentangled on the autoencoder input using efficient signal processing methods instead of bottleneck tuning. Evaluation results show that SpeechSplit 2.0 achieves comparable performance to SpeechS…
Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）；Ping An Technology (Shenzhen) Co., Ltd.（平安科技（深圳）有限公司）；The Chinese University of Hong Kong（香港中文大学）
One-shot voice conversion (VC) with only a single target speaker's speech for reference has become a hot research topic. Existing works generally disentangle timbre, while information about pitch, rhythm and content is still mixed together. To perform one-shot VC effectively with further disentangling these speech components, we employ random resampling for pitch and content encoder and use the variational contrastive log-ratio upper bound of mutual information and gradient reversal layer based adversarial mutual information learning to ensure the different parts of the latent space containing…
One-shot voice conversion (VC), which performs conversion across arbitrary speakers with only a single target-speaker utterance for reference, can be effectively achieved by speech representation disentanglement. Existing work generally ignores the correlation between different speech representations during training, which causes leakage of content information into the speaker representation and thus degrades VC performance. To alleviate this issue, we employ vector quantization (VQ) for content encoding and introduce mutual information (MI) as the correlation metric during training, to achiev…
The notion of disentangled autoencoders was proposed as an extension to the variational autoencoder by introducing a disentanglement parameter $β$, controlling the learning pressure put on the possible underlying latent representations. For certain values of $β$ this kind of autoencoders is capable of encoding independent input generative factors in separate elements of the code, leading to a more interpretable and predictable model behaviour. In this paper we quantify the effects of the parameter $β$ on the model performance and disentanglement. After training multiple models with the same va…
The Chinese University of Hong Kong（香港中文大学）；Centre for Perceptual and Interactive Intelligence, CUHK（香港中文大学感知与交互智能中心）；Tsinghua-CUHK Joint Research Center for Media Sciences, Technologies and Systems（清华-香港中文大学媒体科学、技术与系统联合研究中心）；Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）；Tsinghua University（清华大学）
We propose an unsupervised learning method to disentangle speech into content representation and speaker identity representation. We apply this method to the challenging one-shot cross-lingual voice conversion task to demonstrate the effectiveness of the disentanglement. Inspired by $β$-VAE, we introduce a learning objective that balances between the information captured by the content and speaker representations. In addition, the inductive biases from the architectural design and the training dataset further encourage the desired disentanglement. Both objective and subjective evaluations show…
University of California at Berkeley（加州大学伯克利分校）；Intel AI Lab（英特尔人工智能实验室）
We present a method for converting the voices between a set of speakers. Our method is based on training multiple autoencoder paths, where there is a single speaker-independent encoder and multiple speaker-dependent decoders. The autoencoders are trained with an addition of an adversarial loss which is provided by an auxiliary classifier in order to guide the output of the encoder to be speaker independent. The training of the model is unsupervised in the sense that it does not require collecting the same utterances from the speakers nor does it require time aligning over phonemes. Due to the …
This paper proposes an interesting voice and accent joint conversion approach, which can convert an arbitrary source speaker's voice to a target speaker with non-native accent. This problem is challenging as each target speaker only has training data in native accent and we need to disentangle accent and speaker information in the conversion model training and re-combine them in the conversion stage. In our recognition-synthesis conversion framework, we manage to solve this problem by two proposed tricks. First, we use accent-dependent speech recognizers to obtain bottleneck features for diffe…
University of Toronto（多伦多大学）；Vector Institute（向量研究所）
We decompose the evidence lower bound to show the existence of a term measuring the total correlation between latent variables. We use this to motivate our $β$-TCVAE (Total Correlation Variational Autoencoder), a refinement of the state-of-the-art $β$-VAE objective for learning disentangled representations, requiring no additional hyperparameters during training. We further propose a principled classifier-free measure of disentanglement called the mutual information gap (MIG). We perform extensive quantitative and qualitative experiments, in both restricted and non-restricted settings, and sho…
Non-parallel many-to-many voice conversion remains an interesting but challenging speech processing task. Many style-transfer-inspired methods such as generative adversarial networks (GANs) and variational autoencoders (VAEs) have been proposed. Recently, AutoVC, a conditional autoencoders (CAEs) based method achieved state-of-the-art results by disentangling the speaker identity and speech content using information-constraining bottlenecks, and it achieves zero-shot conversion by swapping in a different speaker's identity embedding to synthesize a new voice. However, we found that while speak…
Tsinghua University（清华大学）；UC San Diego（加州大学圣地亚哥分校）；Qualcomm, Inc.（高通公司）；University of Amsterdam（阿姆斯特丹大学）
We propose an algorithm, guided variational autoencoder (Guided-VAE), that is able to learn a controllable generative model by performing latent representation disentanglement learning. The learning objective is achieved by providing signals to the latent encoding/embedding in VAE without changing its main backbone architecture, hence retaining the desirable properties of the VAE. We design an unsupervised strategy and a supervised strategy in Guided-VAE and observe enhanced modeling and controlling capability over the vanilla VAE. In the unsupervised strategy, we guide the VAE learning by int…
Delft University of Technology（代尔夫特理工大学）；Universitat Politecnica de Catalunya（加泰罗尼亚理工大学）；University of Amsterdam（阿姆斯特丹大学）；Netherlands Cancer Institute（荷兰癌症研究所）；Johns Hopkins University（约翰斯·霍普金斯大学）
In this paper, we propose a new approach to pathological speech synthesis. Instead of using healthy speech as a source, we customise an existing pathological speech sample to a new speaker's voice characteristics. This approach alleviates the evaluation problem one normally has when converting typical speech to pathological speech, as in our approach, the voice conversion (VC) model does not need to be optimised for speech degradation but only for the speaker change. This change in the optimisation ensures that any degradation found in naturalness is due to the conversion process and not due t…
Complex multivariate time series arise in many fields, ranging from computer vision to robotics or medicine. Often we are interested in the independent underlying factors that give rise to the high-dimensional data we are observing. While many models have been introduced to learn such disentangled representations, only few attempt to explicitly exploit the structure of sequential data. We investigate the disentanglement properties of Gaussian process variational autoencoders, a class of models recently introduced that have been successful in different tasks on time series data. Our model explo…
Disentanglement with Factor Quantized Variational Autoencoders
Department of Mathematics and Computer Science, University of Southern Denmark（数学与计算机科学系，丹麦南部大学）；Department of Computer Engineering, Istanbul Technical University（计算机工程系，伊斯坦布尔技术大学）；Department of AI and Data Engineering, Istanbul Technical University（人工智能与数据工程系，伊斯坦布尔技术大学）
This paper presents a low-latency real-time (LLRT) non-parallel voice conversion (VC) framework based on cyclic variational autoencoder (CycleVAE) and multiband WaveRNN with data-driven linear prediction (MWDLP). CycleVAE is a robust non-parallel multispeaker spectral model, which utilizes a speaker-independent latent space and a speaker-dependent code to generate reconstructed/converted spectral features given the spectral features of an input speaker. On the other hand, MWDLP is an efficient and a high-quality neural vocoder that can handle multispeaker data and generate speech waveform for …
This paper presents a refinement framework of WaveNet vocoders for variational autoencoder (VAE) based voice conversion (VC), which reduces the quality distortion caused by the mismatch between the training data and testing data. Conventional WaveNet vocoders are trained with natural acoustic features but conditioned on the converted features in the conversion stage for VC, and such a mismatch often causes significant quality and similarity degradation. In this work, we take advantage of the particular structure of VAEs to refine WaveNet vocoders with the self-reconstructed features generated …
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；MIT-IBM Watson AI Lab（MIT-IBM沃森人工智能实验室）；IBM Research（IBM研究院）
Non-parallel many-to-many voice conversion, as well as zero-shot voice conversion, remain under-explored areas. Deep style transfer algorithms, such as generative adversarial networks (GAN) and conditional variational autoencoder (CVAE), are being applied as new solutions in this field. However, GAN training is sophisticated and difficult, and there is no strong evidence that its generated speech is of good perceptual quality. On the other hand, CVAE training is simple but does not come with the distribution-matching property of a GAN. In this paper, we propose a new style transfer scheme that…
Eindhoven University of Technology（埃因霍温理工大学）；Prosus AI（Prosus人工智能）；Prosus
A disentangled representation of a data set should be capable of recovering the underlying factors that generated it. One question that arises is whether using Euclidean space for latent variable models can produce a disentangled representation when the underlying generating factors have a certain geometrical structure. Take for example the images of a car seen from different angles. The angle has a periodic structure but a 1-dimensional representation would fail to capture this topology. How can we address this problem? The submissions presented for the first stage of the NeurIPS2019 Disentan…
Graduate School of Information Science, Nagoya University（名古屋大学信息科学研究生院）；Information Technology Center, Nagoya University（名古屋大学信息技术中心）
In this paper, we present a description of the baseline system of Voice Conversion Challenge (VCC) 2020 with a cyclic variational autoencoder (CycleVAE) and Parallel WaveGAN (PWG), i.e., CycleVAEPWG. CycleVAE is a nonparallel VAE-based voice conversion that utilizes converted acoustic features to consider cyclically reconstructed spectra during optimization. On the other hand, PWG is a non-autoregressive neural vocoder that is based on a generative adversarial network for a high-quality and fast waveform generator. In practice, the CycleVAEPWG system can be straightforwardly developed with the…
Learning a disentangled, interpretable, and structured latent representation in 3D generative models of faces and bodies is still an open problem. The problem is particularly acute when control over identity features is required. In this paper, we propose an intuitive yet effective self-supervised approach to train a 3D shape variational autoencoder (VAE) which encourages a disentangled latent representation of identity features. Curating the mini-batch generation by swapping arbitrary features across different shapes allows to define a loss function leveraging known differences and similariti…
Variational autoencoders (VAEs) have recently been used for unsupervised disentanglement learning of complex density distributions. Numerous variants exist to encourage disentanglement in latent space while improving reconstruction. However, none have simultaneously managed the trade-off between attaining extremely low reconstruction error and a high disentanglement score. We present a generalized framework to handle this challenge under constrained optimization and demonstrate that it outperforms state-of-the-art existing models as regards disentanglement while balancing reconstruction. We in…
DeNA Co., Ltd.（DeNA有限公司）；University of Tokyo（东京大学）；The University of Electro-Communications（电气通信大学）
Variational autoencoder-based voice conversion (VAE-VC) has the advantage of requiring only pairs of speeches and speaker labels for training. Unlike the majority of the research in VAE-VC which focuses on utilizing auxiliary losses or discretizing latent variables, this paper investigates how an increasing model expressiveness has benefits and impacts on the VAE-VC. Specifically, we first analyze VAE-VC from a rate-distortion perspective, and point out that model expressiveness is significant for VAE-VC because rate and distortion reflect similarity and naturalness of converted speeches. Base…
Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院）；Institute for Artificial Intelligence, KAIST（韩国科学技术院人工智能研究所）
We aim to separate the generative factors of data into two latent vectors in a variational autoencoder. One vector captures class factors relevant to target classification tasks, while the other vector captures style factors relevant to the remaining information. To learn the discrete class features, we introduce supervision using a small amount of labeled data, which can simply yet effectively reduce the effort required for hyperparameter tuning performed in existing unsupervised methods. Furthermore, we introduce a learning objective to encourage statistical independence between the vectors.…
Graduate School of Information Science, Nagoya University（名古屋大学信息科学研究生院）；Information Technology Center, Nagoya University（名古屋大学信息技术中心）
In this paper, we present a novel technique for a non-parallel voice conversion (VC) with the use of cyclic variational autoencoder (CycleVAE)-based spectral modeling. In a variational autoencoder(VAE) framework, a latent space, usually with a Gaussian prior, is used to encode a set of input features. In a VAE-based VC, the encoded latent features are fed into a decoder, along with speaker-coding features, to generate estimated spectra with either the original speaker identity (reconstructed) or another speaker identity (converted). Due to the non-parallel modeling condition, the converted spe…
Variational AutoEncoders (VAEs) provide a means to generate representational latent embeddings. Previous research has highlighted the benefits of achieving representations that are disentangled, particularly for downstream tasks. However, there is some debate about how to encourage disentanglement with VAEs and evidence indicates that existing implementations of VAEs do not achieve disentanglement consistently. The evaluation of how well a VAE's latent space has been disentangled is often evaluated against our subjective expectations of which attributes should be disentangled for a given probl…
We develop a generalisation of disentanglement in VAEs---decomposition of the latent representation---characterising it as the fulfilment of two factors: a) the latent encodings of the data having an appropriate level of overlap, and b) the aggregate encoding of the data conforming to a desired structure, represented through the prior. Decomposition permits disentanglement, i.e. explicit independence between latents, as a special case, but also allows for a much richer class of properties to be imposed on the learnt representation, such as sparsity, clustering, independent subspaces, or even i…
Tsinghua-SPD Bank Joint Research Center for Digital Finance Technologies（清华-浦发银行数字金融技术联合研究中心）
Recent research showed that an autoencoder trained with speech of a single speaker, called exemplar autoencoder (eAE), can be used for any-to-one voice conversion (VC). Compared to large-scale many-to-many models such as AutoVC, the eAE model is easy and fast in training, and may recover more details of the target speaker. To ensure VC quality, the latent code should represent and only represent content information. However, this is not easy to attain for eAE as it is unaware of any speaker variation in model training. To tackle the problem, we propose a simple yet effective approach based on …
Ping An Technology (Shenzhen) Co., Ltd.（平安科技（深圳）有限公司）；University of Science and Technology of China（中国科学技术大学）；Aquinas International Academy（阿奎那国际学院）
Non-parallel many-to-many voice conversion remains an interesting but challenging speech processing task. Recently, AutoVC, a conditional autoencoder based method, achieved excellent conversion results by disentangling the speaker identity and the speech content using information-constraining bottlenecks. However, due to the pure autoencoder training method, it is difficult to evaluate the separation effect of content and speaker identity. In this paper, a novel voice conversion framework, named $\boldsymbol T$ext $\boldsymbol G$uided $\boldsymbol A$utoVC(TGAVC), is proposed to more effectivel…
Singapore University of Technology and Design（新加坡科技设计大学）；University of Southern California（南加州大学）；Institute of High Performance Computing, A*STAR（新加坡科技研究局高性能计算研究所）；Yong Siew Toh Conservatory of Music, National University of Singapore（新加坡国立大学杨秀桃音乐学院）
We propose a flexible framework that deals with both singer conversion and singers vocal technique conversion. The proposed model is trained on non-parallel corpora, accommodates many-to-many conversion, and leverages recent advances of variational autoencoders. It employs separate encoders to learn disentangled latent representations of singer identity and vocal technique separately, with a joint decoder for reconstruction. Conversion is carried out by simple vector arithmetic in the learned latent spaces. Both a quantitative analysis as well as a visualization of the converted spectrograms s…
This paper presents the description of our submitted system for Voice Conversion Challenge (VCC) 2020 with vector-quantization variational autoencoder (VQ-VAE) with WaveNet as the decoder, i.e., VQ-VAE-WaveNet. VQ-VAE-WaveNet is a nonparallel VAE-based voice conversion that reconstructs the acoustic features along with separating the linguistic information with speaker identity. The model is further improved with the WaveNet cycle as the decoder to generate the high-quality speech waveform, since WaveNet, as an autoregressive neural vocoder, has achieved the SoTA result of waveform generation.…
In this paper, we present an open-source software for developing a nonparallel voice conversion (VC) system named crank. Although we have released an open-source VC software based on the Gaussian mixture model named sprocket in the last VC Challenge, it is not straightforward to apply any speech corpus because it is necessary to prepare parallel utterances of source and target speakers to model a statistical conversion function. To address this issue, in this study, we developed a new open-source VC software that enables users to model the conversion function by using only a nonparallel speech…
In this work, we investigate the effectiveness of two techniques for improving variational autoencoder (VAE) based voice conversion (VC). First, we reconsider the relationship between vocoder features extracted using the high quality vocoders adopted in conventional VC systems, and hypothesize that the spectral features are in fact F0 dependent. Such hypothesis implies that during the conversion phase, the latent codes and the converted features in VAE based VC are in fact source F0 dependent. To this end, we propose to utilize the F0 as an additional input of the decoder. The model can learn …
NTT Communication Science Laboratories, NTT Corporation（NTT株式会社NTT通信科学实验室）
This paper proposes a non-parallel many-to-many voice conversion (VC) method using a variant of the conditional variational autoencoder (VAE) called an auxiliary classifier VAE (ACVAE). The proposed method has three key features. First, it adopts fully convolutional architectures to construct the encoder and decoder networks so that the networks can learn conversion rules that capture time dependencies in the acoustic feature sequences of source and target speech. Second, it uses an information-theoretic regularization for the model training to ensure that the information in the attribute clas…
Recent advances in imaging from celestial objects in astronomy visualized via optical and radio telescopes to atoms and molecules resolved via electron and probe microscopes are generating immense volumes of imaging data, containing information about the structure of the universe from atomic to astronomic levels. The classical deep convolutional neural network architectures traditionally perform poorly on the data sets having a significant orientational disorder, that is, having multiple copies of the same or similar object in arbitrary orientation in the image plane. Similarly, while clusteri…
Geometric disentanglement, the separation of latent codes for intrinsic (i.e. identity) and extrinsic(i.e. pose) geometry, is a prominent task for generative models of non-Euclidean data such as 3D deformable models. It provides greater interpretability of the latent space, and leads to more control in generation. This work introduces a mesh feature, the conformal factor and normal feature (CFAN),for use in mesh convolutional autoencoders. We further propose CFAN-VAE, a novel architecture that disentangles identity and pose using the CFAN feature. Requiring no label information on the identity…
Graduate School of Engineering, The University of Tokyo（东京大学工学研究生院）
Recent advances in neural autoregressive models have improve the performance of speech synthesis (SS). However, as they lack the ability to model global characteristics of speech (such as speaker individualities or speaking styles), particularly when these characteristics have not been labeled, making neural autoregressive SS systems more expressive is still an open issue. In this paper, we propose to combine VoiceLoop, an autoregressive SS model, with Variational Autoencoder (VAE). This approach, unlike traditional autoregressive SS systems, uses VAE to model the global characteristics explic…
The task of few-shot style transfer for voice cloning in text-to-speech (TTS) synthesis aims at transferring speaking styles of an arbitrary source speaker to a target speaker's voice using very limited amount of neutral data. This is a very challenging task since the learning algorithm needs to deal with few-shot voice cloning and speaker-prosody disentanglement at the same time. Accelerating the adaptation process for a new target speaker is of importance in real-world applications, but even more challenging. In this paper, we approach to the hard fast few-shot style transfer for voice cloni…
In this paper, we study the disentanglement of speaker and language representations in non-autoregressive cross-lingual TTS models from various aspects. We propose a phoneme length regulator that solves the length mismatch problem between IPA input sequence and monolingual alignment results. Using the phoneme length regulator, we present a FastPitch-based cross-lingual model with IPA symbols as input representations. Our experiments show that language-independent input representations (e.g. IPA symbols), an increasing number of training speakers, and explicit modeling of speech variance inform…
