Neural source-filter (NSF) waveform models generate speech waveforms by morphing sine-based source signals through dilated convolution in the time domain. Although the sine-based source signals help the NSF models to produce voiced sounds with specified pitch, the sine shape may constrain the generated waveform when the target voiced sounds are less periodic. In this paper, we propose a more flexible source signal called cyclic noise, a quasi-periodic noise sequence given by the convolution of a pulse train and a static random noise with a trainable decaying rate that controls the signal shape. We further propose a masked spectral loss to guide the NSF models to produce periodic voiced sounds from the cyclic noise-based source signal. Results from a large-scale listening test demonstrated the effectiveness of the cyclic noise and the masked spectral loss on speaker-independent NSF models in copy-synthesis experiments on the CMU ARCTIC database.
Speech recognition from raw waveform involves learning the spectral decomposition of the signal in the first layer of the neural acoustic model using a convolution layer. In this work, we propose a raw waveform convolutional filter learning approach using soft self-attention. The acoustic filter bank in the proposed model is implemented using a parametric cosine-modulated Gaussian filter bank whose parameters are learned. A network-in-network architecture provides self-attention to generate attention weights over the sub-band filters. The attention weighted log filter bank energies are fed to …
Budapest University of Technology and Economics（布达佩斯技术与经济大学）；MTA-ELTE Lendület Lingual Articulation Research Group（MTA-ELTE Lendület 语言发音研究组）
To date, various speech technology systems have adopted the vocoder approach, a method for synthesizing speech waveform that shows a major role in the performance of statistical parametric speech synthesis. WaveNet one of the best models that nearly resembles the human voice, has to generate a waveform in a time consuming sequential manner with an extremely complex structure of its neural networks.
In this study, we propose an encoder-decoder structured system with fully convolutional networks to implement voice activity detection (VAD) directly on the time-domain waveform. The proposed system processes the input waveform to identify its segments to be either speech or non-speech. This novel waveform-based VAD algorithm, with a short-hand notation "WVAD", has two main particularities. First, as compared to most conventional VAD systems that use spectral features, raw-waveforms employed in WVAD contain more comprehensive information and thus are supposed to facilitate more accurate speech…
National Institute of Informatics（国立信息学研究所）；The University of Electro-Communications（电气通信大学）；The University of Edinburgh（爱丁堡大学）
This paper proposes a new loss using short-time Fourier transform (STFT) spectra for the aim of training a high-performance neural speech waveform model that predicts raw continuous speech waveform samples directly. Not only amplitude spectra but also phase spectra obtained from generated speech waveforms are used to calculate the proposed loss. We also mathematically show that training of the waveform model on the basis of the proposed loss can be interpreted as maximum likelihood training that assumes the amplitude and phase spectra of generated speech waveforms following Gaussian and von Mi…
Recently, GAN based speech synthesis methods, such as MelGAN, have become very popular. Compared to conventional autoregressive based methods, parallel structures based generators make waveform generation process fast and stable. However, the quality of generated speech by autoregressive based neural vocoders, such as WaveRNN, is still higher than GAN. To address this issue, we propose a novel vocoder model: TFGAN, which is adversarially learned both in time and frequency domain. On one hand, we propose to discriminate ground-truth waveform from synthetic one in frequency domain for offering m…
Statistical TTS systems that directly predict the speech waveform have recently reported improvements in synthesis quality. This investigation evaluates Amazon's statistical speech waveform synthesis (SSWS) system. An in-depth evaluation of SSWS is conducted across a number of domains to better understand the consistency in quality. The results of this evaluation are validated by repeating the procedure on a separate group of testers. Finally, an analysis of the nature of speech errors of SSWS compared to hybrid unit selection synthesis is conducted to identify the strengths and weaknesses of …
University of Science and Technology of China（中国科学技术大学）；Baidu Speech Department（百度语音部门）
This paper presents a waveform modeling and generation method using hierarchical recurrent neural networks (HRNN) for speech bandwidth extension (BWE). Different from conventional BWE methods which predict spectral parameters for reconstructing wideband speech waveforms, this BWE method models and predicts waveform samples directly without using vocoders. Inspired by SampleRNN which is an unconditional neural audio generator, the HRNN model represents the distribution of each wideband or high-frequency waveform sample conditioned on the input narrowband waveform samples using a neural network …
NTT Communication Science Laboratories, NTT Corporation（NTT公司NTT通信科学实验室）
WaveCycleGAN has recently been proposed to bridge the gap between natural and synthesized speech waveforms in statistical parametric speech synthesis and provides fast inference with a moving average model rather than an autoregressive model and high-quality speech synthesis with the adversarial training. However, the human ear can still distinguish the processed speech waveforms from natural ones. One possible cause of this distinguishability is the aliasing observed in the processed speech waveform via down/up-sampling modules. To solve the aliasing and provide higher quality speech synthesi…
This study investigates the direct use of speech waveforms to predict head motion for speech-driven head-motion synthesis, whereas the use of spectral features such as MFCC as basic input features together with additional features such as energy and F0 is common in the literature. We show that, rather than combining different features that originate from waveforms, it is more effective to use waveforms directly predicting corresponding head motion. The challenge with the waveform-based approach is that waveforms contain a large amount of information irrelevant to predict head motion, which hin…
NTT Communication Science Laboratories, NTT Corporation（NTT公司NTT通信科学实验室）
We propose a learning-based filter that allows us to directly modify a synthetic speech waveform into a natural speech waveform. Speech-processing systems using a vocoder framework such as statistical parametric speech synthesis and voice conversion are convenient especially for a limited number of data because it is possible to represent and process interpretable acoustic features over a compact space, such as the fundamental frequency (F0) and mel-cepstrum. However, a well-known problem that leads to the quality degradation of generated speech is an over-smoothing effect that eliminates some…
Distinguishing Neural Speech Synthesis Models Through Fingerprints in Speech Waveforms
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Tsinghua University（清华大学）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）
National Institute of Informatics（国立信息学研究所）；Massachusetts Institute of Technology（麻省理工学院）；University of Edinburgh（爱丁堡大学）
Vector Quantized Variational AutoEncoders (VQ-VAE) are a powerful representation learning framework that can discover discrete groups of features from a speech signal without supervision. Until now, the VQ-VAE architecture has previously modeled individual types of speech features, such as only phones or only F0. This paper introduces an important extension to VQ-VAE for learning F0-related suprasegmental information simultaneously along with traditional phone features.The proposed framework uses two encoders such that the F0 trajectory and speech waveform are both input to the system, therefo…
We propose PeriodNet, a non-autoregressive (non-AR) waveform generation model with a new model structure for modeling periodic and aperiodic components in speech waveforms. The non-AR waveform generation models can generate speech waveforms parallelly and can be used as a speech vocoder by conditioning an acoustic feature. Since a speech waveform contains periodic and aperiodic components, both components should be appropriately modeled to generate a high-quality speech waveform. However, it is difficult to decompose the components from a natural speech waveform in advance. To address this iss…
Neural waveform models such as the WaveNet are used in many recent text-to-speech systems, but the original WaveNet is quite slow in waveform generation because of its autoregressive (AR) structure. Although faster non-AR models were recently reported, they may be prohibitively complicated due to the use of a distilling training method and the blend of other disparate training criteria. This study proposes a non-AR neural source-filter waveform model that can be directly trained using spectrum-based training criteria and the stochastic gradient descent method. Given the input acoustic features…
In this work, we present CleanUNet, a causal speech denoising model on the raw waveform. The proposed model is based on an encoder-decoder architecture combined with several self-attention blocks to refine its bottleneck representations, which is crucial to obtain good results. The model is optimized through a set of losses defined over both waveform and multi-resolution spectrograms. The proposed method outperforms the state-of-the-art models in terms of denoised speech quality from various objective and subjective evaluation metrics. We release our code and models at https://github.com/nvidi…
In real acoustic environment, speech enhancement is an arduous task to improve the quality and intelligibility of speech interfered by background noise and reverberation. Over the past years, deep learning has shown great potential on speech enhancement. In this paper, we propose a novel real-time framework called DBNet which is a dual-branch structure with alternate interconnection. Each branch incorporates an encoder-decoder architecture with skip connections. The two branches are responsible for spectrum and waveform modeling, respectively. A bridge layer is adopted to exchange information …
The performance of text-to-speech (TTS) systems heavily depends on spectrogram to waveform generation, also known as the speech reconstruction phase. The time required for the same is known as synthesis delay. In this paper, an approach to reduce speech synthesis delay has been proposed. It aims to enhance the TTS systems for real-time applications such as digital assistants, mobile phones, embedded devices, etc. The proposed approach applies Fast Griffin Lim Algorithm (FGLA) instead Griffin Lim algorithm (GLA) as vocoder in the speech synthesis phase. GLA and FGLA are both iterative, but the …
National Institute of Informatics（国立信息学研究所）；SOKENDAI（综合研究大学院大学）；University of Science and Technology of China（中国科学技术大学）
High-quality speech corpora are essential foundations for most speech applications. However, such speech data are expensive and limited since they are collected in professional recording environments. In this work, we propose an encoder-decoder neural network to automatically enhance low-quality recordings to professional high-quality recordings. To address channel variability, we first filter out the channel characteristics from the original input audio using the encoder network with adversarial training. Next, we disentangle the channel factor from a reference audio. Conditioned on this fact…
National Institute of Informatics, Tokyo, Japan（日本东京国立信息学研究所）；Department of Signal Processing and Acoustics, Aalto University, Finland（芬兰阿尔托大学信号处理与声学系）
Recent neural waveform synthesizers such as WaveNet, WaveGlow, and the neural-source-filter (NSF) model have shown good performance in speech synthesis despite their different methods of waveform generation. The similarity between speech and music audio synthesis techniques suggests interesting avenues to explore in terms of the best way to apply speech synthesizers in the music domain. This work compares three neural synthesizers used for musical instrument sounds generation under three scenarios: training from scratch on music data, zero-shot learning from the speech domain, and fine-tuning-…
Aalto University（阿尔托大学）；National Institute of Informatics（国立信息学研究所）；University of Edinburgh（爱丁堡大学）
The state-of-the-art in text-to-speech synthesis has recently improved considerably due to novel neural waveform generation methods, such as WaveNet. However, these methods suffer from their slow sequential inference process, while their parallel versions are difficult to train and even more expensive computationally. Meanwhile, generative adversarial networks (GANs) have achieved impressive results in image generation and are making their way into audio applications; parallel inference is among their lucrative properties. By adopting recent advances in GAN training techniques, this investigat…
National Institute of Informatics（国立信息学研究所）；University of Edinburgh（爱丁堡大学）
Neural source-filter (NSF) models are deep neural networks that produce waveforms given input acoustic features. They use dilated-convolution-based neural filter modules to filter sine-based excitation for waveform generation, which is different from WaveNet and flow-based models. One of the NSF models, called harmonic-plus-noise NSF (h-NSF) model, uses separate pairs of source and neural filters to generate harmonic and noise waveform components. It is close to WaveNet in terms of speech quality while being superior in generation speed. The h-NSF model can be improved even further. While h-NS…
Recent studies have shown that text-to-speech synthesis quality can be improved by using glottal vocoding. This refers to vocoders that parameterize speech into two parts, the glottal excitation and vocal tract, that occur in the human speech production apparatus. Current glottal vocoders generate the glottal excitation waveform by using deep neural networks (DNNs). However, the squared error-based training of the present glottal excitation models is limited to generating conditional average waveforms, which fails to capture the stochastic variation of the waveforms. As a result, shaped noise …
Aalto University（阿尔托大学）；University of Crete（克里特大学）；National Institute of Informatics（国立信息学研究所）
Recent speech technology research has seen a growing interest in using WaveNets as statistical vocoders, i.e., generating speech waveforms from acoustic features. These models have been shown to improve the generated speech quality over classical vocoders in many tasks, such as text-to-speech synthesis and voice conversion. Furthermore, conditioning WaveNets with acoustic features allows sharing the waveform generator model across multiple speakers without additional speaker codes. However, multi-speaker WaveNet models require large amounts of training data and computation to cover the entire …
In this paper, we propose multi-band MelGAN, a much faster waveform generation model targeting to high-quality text-to-speech. Specifically, we improve the original MelGAN by the following aspects. First, we increase the receptive field of the generator, which is proven to be beneficial to speech generation. Second, we substitute the feature matching loss with the multi-resolution STFT loss to better measure the difference between fake and real speech. Together with pre-training, this improvement leads to both better quality and better training stability. More importantly, we extend MelGAN wit…
National Taiwan University（台湾大学）；Academia Sinica（台湾中央研究院）；National Cheng-Kung University（台湾成功大学）
In recent years, waveform-mapping-based speech enhancement (SE) methods have garnered significant attention. These methods generally use a deep learning model to directly process and reconstruct speech waveforms. Because both the input and output are in waveform format, the waveform-mapping-based SE methods can overcome the distortion caused by imperfect phase estimation, which may be encountered in spectral-mapping-based SE systems. So far, most waveform-mapping-based SE methods have focused on single-channel tasks. In this paper, we propose a novel fully convolutional network (FCN) with Sinc…
Beijing University of Posts and Telecommunications（北京邮电大学）；Peng Cheng Laboratory（鹏城实验室）
We propose a novel neural waveform compression method to catalyze emerging speech semantic communications. By introducing nonlinear transform and variational modeling, we effectively capture the dependencies within speech frames and estimate the probabilistic distribution of the speech feature more accurately, giving rise to better compression performance. In particular, the speech signals are analyzed and synthesized by a pair of nonlinear transforms, yielding latent features. An entropy model with hyperprior is built to capture the probabilistic distribution of latent features, followed with…
This paper presents a neural vocoder named HiNet which reconstructs speech waveforms from acoustic features by predicting amplitude and phase spectra hierarchically. Different from existing neural vocoders such as WaveNet, SampleRNN and WaveRNN which directly generate waveform samples using single neural networks, the HiNet vocoder is composed of an amplitude spectrum predictor (ASP) and a phase spectrum predictor (PSP). The ASP is a simple DNN model which predicts log amplitude spectra (LAS) from acoustic features. The predicted LAS are sent into the PSP for phase recovery. Considering the is…
This paper proposes an effective probability density distillation (PDD) algorithm for WaveNet-based parallel waveform generation (PWG) systems. Recently proposed teacher-student frameworks in the PWG system have successfully achieved a real-time generation of speech signals. However, the difficulties optimizing the PDD criteria without auxiliary losses result in quality degradation of synthesized speech. To generate more natural speech signals within the teacher-student framework, we propose a novel optimization criterion based on generative adversarial networks (GANs). In the proposed method,…
Speech recognition in noisy and channel distorted scenarios is often challenging as the current acoustic modeling schemes are not adaptive to the changes in the signal distribution in the presence of noise. In this work, we develop a novel acoustic modeling framework for noise robust speech recognition based on relevance weighting mechanism. The relevance weighting is achieved using a sub-network approach that performs feature selection. A relevance sub-network is applied on the output of first layer of a convolutional network model operating on raw speech signals while a second relevance sub-…
Facebook AI Research（脸书人工智能研究院）；INRIA（法国国家信息与自动化研究所）；PSL Research University（巴黎文理研究大学）
We present a causal speech enhancement model working on the raw waveform that runs in real-time on a laptop CPU. The proposed model is based on an encoder-decoder architecture with skip-connections. It is optimized on both time and frequency domains, using multiple loss functions. Empirical evidence shows that it is capable of removing various kinds of background noise including stationary and non-stationary noises, as well as room reverb. Additionally, we suggest a set of data augmentation techniques applied directly on the raw waveform which further improve model performance and its generali…
National Institute of Informatics（国立信息学研究所）；Aalto University（阿尔托大学）
Recent advances in speech synthesis suggest that limitations such as the lossy nature of the amplitude spectrum with minimum phase approximation and the over-smoothing effect in acoustic modeling can be overcome by using advanced machine learning approaches. In this paper, we build a framework in which we can fairly compare new vocoding and acoustic modeling techniques with conventional approaches by means of a large scale crowdsourced evaluation. Results on acoustic models showed that generative adversarial networks and an autoregressive (AR) model performed better than a normal recurrent net…
Shanghai University（上海大学）；Shanghai Institute for Advanced Communication and Data Science (SICS)（上海先进通信与数据科学研究院（SICS））
In recent years, synthetic speech generated by advanced text-to-speech (TTS) and voice conversion (VC) systems has caused great harms to automatic speaker verification (ASV) systems, urging us to design a synthetic speech detection system to protect ASV systems. In this paper, we propose a new speech anti-spoofing model named ResWavegram-Resnet (RW-Resnet). The model contains two parts, Conv1D Resblocks and backbone Resnet34. The Conv1D Resblock is based on the Conv1D block with a residual connection. For the first part, we use the raw waveform as input and feed it to the stacked Conv1D Resblo…
Aalto University（阿尔托大学）；National Institute of Informatics（国立信息学研究所）；NTT Communication Science Laboratories, NTT Corporation（NTT公司NTT通信科学实验室）
This paper proposes a method for generating speech from filterbank mel frequency cepstral coefficients (MFCC), which are widely used in speech applications, such as ASR, but are generally considered unusable for speech synthesis. First, we predict fundamental frequency and voicing information from MFCCs with an autoregressive recurrent neural net. Second, the spectral envelope information contained in MFCCs is converted to all-pole filters, and a pitch-synchronous excitation model matched to these filters is trained. Finally, we introduce a generative adversarial network -based noise model to …
Neural-based text-to-speech (TTS) systems achieve very high-fidelity speech generation because of the rapid neural network developments. However, the huge labeled corpus and high computation cost requirements limit the possibility of developing a high-fidelity TTS system by small companies or individuals. On the other hand, a neural vocoder, which has been widely adopted for the speech generation in neural-based TTS systems, can be trained with a relatively small unlabeled corpus. Therefore, in this paper, we explore a general framework to develop a neural post-filter (NPF) for low-cost TTS sy…
End-to-end speech recognition generally uses hand-engineered acoustic features as input and excludes the feature extraction module from its joint optimization. To extract learnable and adaptive features and mitigate information loss, we propose a new encoder that adopts globally attentive locally recurrent (GALR) networks and directly takes raw waveform as input. We observe improved ASR performance and robustness by applying GALR on different window lengths to aggregate fine-grain temporal information into multi-scale acoustic features. Experiments are conducted on a benchmark dataset AISHELL-…
Neural vocoders based on the generative adversarial neural network (GAN) have been widely used due to their fast inference speed and lightweight networks while generating high-quality speech waveforms. Since the perceptually important speech components are primarily concentrated in the low-frequency bands, most GAN-based vocoders perform multi-scale analysis that evaluates downsampled speech waveforms. This multi-scale analysis helps the generator improve speech intelligibility. However, in preliminary experiments, we discovered that the multi-scale analysis which focuses on the low-frequency …
National Institute of Informatics（国立信息学研究所）；NTT Communication Science Laboratories, Nippon Telegraph and Telephone Corporation（日本电报电话公司NTT通信科学实验室）；University of Edinburgh（爱丁堡大学）
Recently, we proposed short-time Fourier transform (STFT)-based loss functions for training a neural speech waveform model. In this paper, we generalize the above framework and propose a training scheme for such models based on spectral amplitude and phase losses obtained by either STFT or continuous wavelet transform (CWT), or both of them. Since CWT is capable of having time and frequency resolutions different from those of STFT and is cable of considering those closer to human auditory scales, the proposed loss functions could provide complementary information on speech signals. Experimenta…
Deep neural networks can learn complex and abstract representations, that are progressively obtained by combining simpler ones. A recent trend in speech and speaker recognition consists in discovering these representations starting from raw audio samples directly. Differently from standard hand-crafted features such as MFCCs or FBANK, the raw waveform can potentially help neural networks discover better and more customized representations. The high-dimensional raw inputs, however, can make training significantly more challenging. This paper summarizes our recent efforts to develop a neural arc…
In recent years, speech enhancement (SE) has achieved impressive progress with the success of deep neural networks (DNNs). However, the DNN approach usually fails to generalize well to unseen environmental noise that is not included in the training. To address this problem, we propose "noise tokens" (NTs), which are a set of neural noise templates that are jointly trained with the SE system. NTs dynamically capture the environment variability and thus enable the DNN model to handle various environments to produce STFT magnitude with higher quality. Experimental results show that using NTs is a…
An emerging trend in audio processing is capturing low-level speech representations from raw waveforms. These representations have shown promising results on a variety of tasks, such as speech recognition and speech separation. Compared to handcrafted features, learning speech features via backpropagation provides the model greater flexibility in how it represents data for different tasks theoretically. However, results from empirical study shows that, in some tasks, such as voice spoof detection, handcrafted features are more competitive than learned features. Instead of evaluating handcrafte…
Speech separation has been studied widely for single-channel close-talk microphone recordings over the past few years; developed solutions are mostly in frequency-domain. Recently, a raw audio waveform separation network (TasNet) is introduced for single-channel data, with achieving high Si-SNR (scale-invariant source-to-noise ratio) and SDR (source-to-distortion ratio) comparing against the state-of-the-art solution in frequency-domain. In this study, we incorporate effective components of the TasNet into a frequency-domain separation method. We compare both for alternative scenarios. We intr…
National Institute of Informatics（国立信息学研究所）；Nagoya Institute of Technology（名古屋工业大学）；Centre for Speech Technology Research, University of Edinburgh（爱丁堡大学语音技术研究中心）
Neural waveform models such as WaveNet have demonstrated better performance than conventional vocoders for statistical parametric speech synthesis. As an autoregressive (AR) model, WaveNet is limited by a slow sequential waveform generation process. Some new models that use the inverse-autoregressive flow (IAF) can generate a whole waveform in a one-shot manner. However, these IAF-based models require sequential transformation during training, which severely slows down the training speed. Other models such as Parallel WaveNet and ClariNet bring together the benefits of AR and IAF-based models …
Imperial College London（伦敦帝国理工学院）；Samsung AI Centre Cambridge（三星剑桥人工智能中心）；University of Augsburg（奥格斯堡大学）；Facebook London（脸书伦敦）
Video-to-speech is the process of reconstructing the audio speech from a video of a spoken utterance. Previous approaches to this task have relied on a two-step process where an intermediate representation is inferred from the video, and is then decoded into waveform audio using a vocoder or a waveform reconstruction algorithm. In this work, we propose a new end-to-end video-to-speech model based on Generative Adversarial Networks (GANs) which translates spoken video to waveform end-to-end without using any intermediate representation or separate waveform synthesis algorithm. Our model consist…
