arXiv subjects

Junichi Yamagishi

Publications and source records attributed to Junichi Yamagishi.

At least 199 records · Page 11Linked to original sources

arXiv2019

Training Multi-Speaker Neural Text-to-Speech Systems using Speaker-Imbalanced Speech Corpora

When the available data of a target speaker is insufficient to train a high quality speaker-dependent neural text-to-speech (TTS) system, we can combine data from multiple speakers and train a multi-speaker TTS model instead. Many studies have shown that neural multi-speaker TTS model trained with a small amount data from multiple speakers combined can generate synthetic speech with better quality and stability than a speaker-dependent one. However when the amount of data from each speaker is highly unbalanced, the best approach to make use of the excessive data remains unknown. Our experiments showed that simply combining all available data from every speaker to train a multi-speaker model produces better than or at least similar performance to its speaker-dependent counterpart. Moreover by using an ensemble multi-speaker model, in which each subsystem is trained on a subset of available data, we can further improve the quality of the synthetic speech especially for underrepresented speakers whose training data is limited.

eess.AS↗

arXiv2019

Investigation of enhanced Tacotron text-to-speech synthesis systems with self-attention for pitch accent language

End-to-end speech synthesis is a promising approach that directly converts raw text to speech. Although it was shown that Tacotron2 outperforms classical pipeline systems with regards to naturalness in English, its applicability to other languages is still unknown. Japanese could be one of the most difficult languages for which to achieve end-to-end speech synthesis, largely due to its character diversity and pitch accents. Therefore, state-of-the-art systems are still based on a traditional pipeline framework that requires a separate text analyzer and duration model. Towards end-to-end Japanese speech synthesis, we extend Tacotron to systems with self-attention to capture long-term dependencies related to pitch accents and compare their audio quality with classical pipeline systems under various conditions to show their pros and cons. In a large-scale listening test, we investigated the impacts of the presence of accentual-type labels, the use of force or predicted alignments, and acoustic features used as local condition parameters of the Wavenet vocoder. Our results reveal that although the proposed systems still do not match the quality of a top-line pipeline system for Japanese, we show important stepping stones towards end-to-end Japanese speech synthesis.

eess.AS↗

arXiv2019

Non linear time compression of clear and normal speech at high rates

We compare a series of time compression methods applied to normal and clear speech. First we evaluate a linear (uniform) method applied to these styles as well as to naturally-produced fast speech. We found, in line with the literature, that unprocessed fast speech was less intelligible than linearly compressed normal speech. Fast speech was also less intelligible than compressed clear speech but at the highest rate (three times faster than normal) the advantage of clear over fast speech was lost. To test whether this was due to shorter speech duration we evaluate, in our second experiments, a range of methods that compress speech and silence at different rates. We found that even when the overall duration of speech and silence is kept the same across styles, compressed normal speech is still more intelligible than compressed clear speech. Compressing silence twice as much as speech improved results further for normal speech with very little additional computational costs.

eess.AS↗

arXiv2019

Introduction to Voice Presentation Attack Detection and Recent Advances

Over the past few years significant progress has been made in the field of presentation attack detection (PAD) for automatic speaker recognition (ASV). This includes the development of new speech corpora, standard evaluation protocols and advancements in front-end feature extraction and back-end classifiers. The use of standard databases and evaluation protocols has enabled for the first time the meaningful benchmarking of different PAD solutions. This chapter summarises the progress, with a focus on studies completed in the last three years. The article presents a summary of findings and lessons learned from two ASVspoof challenges, the first community-led benchmarking efforts. These show that ASV PAD remains an unsolved problem and that further attention is required to develop generalised PAD solutions which have potential to detect diverse and previously unseen spoofing attacks.

cs.SD↗

arXiv2018

Identifying Computer-Translated Paragraphs using Coherence Features

We have developed a method for extracting the coherence features from a paragraph by matching similar words in its sentences. We conducted an experiment with a parallel German corpus containing 2000 human-created and 2000 machine-translated paragraphs. The result showed that our method achieved the best performance (accuracy = 72.3%, equal error rate = 29.8%) when it is compared with previous methods on various computer-generated text including translation and paper generation (best accuracy = 67.9%, equal error rate = 32.0%). Experiments on Dutch, another rich resource language, and a low resource one (Japanese) attained similar performances. It demonstrated the efficiency of the coherence features at distinguishing computer-translated from human-created paragraphs on diverse languages.

cs.CL↗

arXiv2018

Audiovisual speaker conversion: jointly and simultaneously transforming facial expression and acoustic characteristics

An audiovisual speaker conversion method is presented for simultaneously transforming the facial expressions and voice of a source speaker into those of a target speaker. Transforming the facial and acoustic features together makes it possible for the converted voice and facial expressions to be highly correlated and for the generated target speaker to appear and sound natural. It uses three neural networks: a conversion network that fuses and transforms the facial and acoustic features, a waveform generation network that produces the waveform from both the converted facial and acoustic features, and an image reconstruction network that outputs an RGB facial image also based on both the converted features. The results of experiments using an emotional audiovisual database showed that the proposed method achieved significantly higher naturalness compared with one that separately transformed acoustic and facial features.

eess.AS↗

arXiv2018

Attentive Filtering Networks for Audio Replay Attack Detection

An attacker may use a variety of techniques to fool an automatic speaker verification system into accepting them as a genuine user. Anti-spoofing methods meanwhile aim to make the system robust against such attacks. The ASVspoof 2017 Challenge focused specifically on replay attacks, with the intention of measuring the limits of replay attack detection as well as developing countermeasures against them. In this work, we propose our replay attacks detection system - Attentive Filtering Network, which is composed of an attention-based filtering mechanism that enhances feature representations in both the frequency and time domains, and a ResNet-based classifier. We show that the network enables us to visualize the automatically acquired feature representations that are helpful for spoofing detection. Attentive Filtering Network attains an evaluation EER of 8.99$\%$ on the ASVspoof 2017 Version 2.0 dataset. With system fusion, our best system further obtains a 30$\%$ relative improvement over the ASVspoof 2017 enhanced baseline system.

eess.AS↗

arXiv2018

STFT spectral loss for training a neural speech waveform model

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 Mises distributions, respectively. Furthermore, this paper presents a simple network architecture as the speech waveform model, which is composed of uni-directional long short-term memories (LSTMs) and an auto-regressive structure. Experimental results showed that the proposed neural model synthesized high-quality speech waveforms.

eess.AS↗

arXiv2018

Waveform generation for text-to-speech synthesis using pitch-synchronous multi-scale generative adversarial networks

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 investigation studies waveform generation for TTS in two domains (speech signal and glottal excitation). Listening test results show that while direct waveform generation with GAN is still far behind WaveNet, a GAN-based glottal excitation model can achieve quality and voice similarity on par with a WaveNet vocoder.

eess.AS↗

arXiv2018

Capsule-Forensics: Using Capsule Networks to Detect Forged Images and Videos

Recent advances in media generation techniques have made it easier for attackers to create forged images and videos. State-of-the-art methods enable the real-time creation of a forged version of a single video obtained from a social network. Although numerous methods have been developed for detecting forged images and videos, they are generally targeted at certain domains and quickly become obsolete as new kinds of attacks appear. The method introduced in this paper uses a capsule network to detect various kinds of spoofs, from replay attacks using printed images or recorded videos to computer-generated videos using deep convolutional neural networks. It extends the application of capsule networks beyond their original intention to the solving of inverse graphics problems.

cs.CV↗

arXiv2018

Scaling and bias codes for modeling speaker-adaptive DNN-based speech synthesis systems

Most neural-network based speaker-adaptive acoustic models for speech synthesis can be categorized into either layer-based or input-code approaches. Although both approaches have their own pros and cons, most existing works on speaker adaptation focus on improving one or the other. In this paper, after we first systematically overview the common principles of neural-network based speaker-adaptive models, we show that these approaches can be represented in a unified framework and can be generalized further. More specifically, we introduce the use of scaling and bias codes as generalized means for speaker-adaptive transformation. By utilizing these codes, we can create a more efficient factorized speaker-adaptive model and capture advantages of both approaches while reducing their disadvantages. The experiments show that the proposed method can improve the performance of speaker adaptation compared with speaker adaptation based on the conventional input code.

eess.AS↗

arXiv2018

Transforming acoustic characteristics to deceive playback spoofing countermeasures of speaker verification systems

Automatic speaker verification (ASV) systems use a playback detector to filter out playback attacks and ensure verification reliability. Since current playback detection models are almost always trained using genuine and played-back speech, it may be possible to degrade their performance by transforming the acoustic characteristics of the played-back speech close to that of the genuine speech. One way to do this is to enhance speech "stolen" from the target speaker before playback. We tested the effectiveness of a playback attack using this method by using the speech enhancement generative adversarial network to transform acoustic characteristics. Experimental results showed that use of this "enhanced stolen speech" method significantly increases the equal error rates for the baseline used in the ASVspoof 2017 challenge and for a light convolutional neural network-based method. The results also showed that its use degrades the performance of a Gaussian mixture model-universal background model-based ASV system. This type of attack is thus an urgent problem needing to be solved.

cs.SD↗

arXiv2018

Deep Encoder-Decoder Models for Unsupervised Learning of Controllable Speech Synthesis

Generating versatile and appropriate synthetic speech requires control over the output expression separate from the spoken text. Important non-textual speech variation is seldom annotated, in which case output control must be learned in an unsupervised fashion. In this paper, we perform an in-depth study of methods for unsupervised learning of control in statistical speech synthesis. For example, we show that popular unsupervised training heuristics can be interpreted as variational inference in certain autoencoder models. We additionally connect these models to VQ-VAEs, another, recently-proposed class of deep variational autoencoders, which we show can be derived from a very similar mathematical argument. The implications of these new probabilistic interpretations are discussed. We illustrate the utility of the various approaches with an application to acoustic modelling for emotional speech synthesis, where the unsupervised methods for learning expression control (without access to emotional labels) are found to give results that in many aspects match or surpass the previous best supervised approach.

eess.AS↗

arXiv2018

A Spoofing Benchmark for the 2018 Voice Conversion Challenge: Leveraging from Spoofing Countermeasures for Speech Artifact Assessment

Voice conversion (VC) aims at conversion of speaker characteristic without altering content. Due to training data limitations and modeling imperfections, it is difficult to achieve believable speaker mimicry without introducing processing artifacts; performance assessment of VC, therefore, usually involves both speaker similarity and quality evaluation by a human panel. As a time-consuming, expensive, and non-reproducible process, it hinders rapid prototyping of new VC technology. We address artifact assessment using an alternative, objective approach leveraging from prior work on spoofing countermeasures (CMs) for automatic speaker verification. Therein, CMs are used for rejecting `fake' inputs such as replayed, synthetic or converted speech but their potential for automatic speech artifact assessment remains unknown. This study serves to fill that gap. As a supplement to subjective results for the 2018 Voice Conversion Challenge (VCC'18) data, we configure a standard constant-Q cepstral coefficient CM to quantify the extent of processing artifacts. Equal error rate (EER) of the CM, a confusability index of VC samples with real human speech, serves as our artifact measure. Two clusters of VCC'18 entries are identified: low-quality ones with detectable artifacts (low EERs), and higher quality ones with less artifacts. None of the VCC'18 systems, however, is perfect: all EERs are < 30 % (the `ideal' value would be 50 %). Our preliminary findings suggest potential of CMs outside of their original application, as a supplemental optimization and benchmarking tool to enhance VC technology.

eess.AS↗

arXiv2018

MesoNet: a Compact Facial Video Forgery Detection Network

This paper presents a method to automatically and efficiently detect face tampering in videos, and particularly focuses on two recent techniques used to generate hyper-realistic forged videos: Deepfake and Face2Face. Traditional image forensics techniques are usually not well suited to videos due to the compression that strongly degrades the data. Thus, this paper follows a deep learning approach and presents two networks, both with a low number of layers to focus on the mesoscopic properties of images. We evaluate those fast networks on both an existing dataset and a dataset we have constituted from online videos. The tests demonstrate a very successful detection rate with more than 98% for Deepfake and 95% for Face2Face.

cs.CV↗

arXiv2018

Multimodal speech synthesis architecture for unsupervised speaker adaptation

This paper proposes a new architecture for speaker adaptation of multi-speaker neural-network speech synthesis systems, in which an unseen speaker's voice can be built using a relatively small amount of speech data without transcriptions. This is sometimes called "unsupervised speaker adaptation". More specifically, we concatenate the layers to the audio inputs when performing unsupervised speaker adaptation while we concatenate them to the text inputs when synthesizing speech from text. Two new training schemes for the new architecture are also proposed in this paper. These training schemes are not limited to speech synthesis, other applications are suggested. Experimental results show that the proposed model not only enables adaptation to unseen speakers using untranscribed speech but it also improves the performance of multi-speaker modeling and speaker adaptation using transcribed audio files.

eess.AS↗

arXiv2018

Investigating accuracy of pitch-accent annotations in neural network-based speech synthesis and denoising effects

We investigated the impact of noisy linguistic features on the performance of a Japanese speech synthesis system based on neural network that uses WaveNet vocoder. We compared an ideal system that uses manually corrected linguistic features including phoneme and prosodic information in training and test sets against a few other systems that use corrupted linguistic features. Both subjective and objective results demonstrate that corrupted linguistic features, especially those in the test set, affected the ideal system's performance significantly in a statistical sense due to a mismatched condition between the training and test sets. Interestingly, while an utterance-level Turing test showed that listeners had a difficult time differentiating synthetic speech from natural speech, it further indicated that adding noise to the linguistic features in the training set can partially reduce the effect of the mismatch, regularize the model, and help the system perform better when linguistic features of the test set are noisy.

eess.AS↗

arXiv2018

Wasserstein GAN and Waveform Loss-based Acoustic Model Training for Multi-speaker Text-to-Speech Synthesis Systems Using a WaveNet Vocoder

Recent neural networks such as WaveNet and sampleRNN that learn directly from speech waveform samples have achieved very high-quality synthetic speech in terms of both naturalness and speaker similarity even in multi-speaker text-to-speech synthesis systems. Such neural networks are being used as an alternative to vocoders and hence they are often called neural vocoders. The neural vocoder uses acoustic features as local condition parameters, and these parameters need to be accurately predicted by another acoustic model. However, it is not yet clear how to train this acoustic model, which is problematic because the final quality of synthetic speech is significantly affected by the performance of the acoustic model. Significant degradation happens, especially when predicted acoustic features have mismatched characteristics compared to natural ones. In order to reduce the mismatched characteristics between natural and generated acoustic features, we propose frameworks that incorporate either a conditional generative adversarial network (GAN) or its variant, Wasserstein GAN with gradient penalty (WGAN-GP), into multi-speaker speech synthesis that uses the WaveNet vocoder. We also extend the GAN frameworks and use the discretized mixture logistic loss of a well-trained WaveNet in addition to mean squared error and adversarial losses as parts of objective functions. Experimental results show that acoustic models trained using the WGAN-GP framework using back-propagated discretized-mixture-of-logistics (DML) loss achieves the highest subjective evaluation scores in terms of both quality and speaker similarity.

eess.AS↗