arXiv subjects

Shinji Watanabe

Publications and source records attributed to Shinji Watanabe.

At least 487 records · Page 27Linked to original sources

arXiv2019

Stream attention-based multi-array end-to-end speech recognition

Automatic Speech Recognition (ASR) using multiple microphone arrays has achieved great success in the far-field robustness. Taking advantage of all the information that each array shares and contributes is crucial in this task. Motivated by the advances of joint Connectionist Temporal Classification (CTC)/attention mechanism in the End-to-End (E2E) ASR, a stream attention-based multi-array framework is proposed in this work. Microphone arrays, acting as information streams, are activated by separate encoders and decoded under the instruction of both CTC and attention networks. In terms of attention, a hierarchical structure is adopted. On top of the regular attention networks, stream attention is introduced to steer the decoder toward the most informative encoders. Experiments have been conducted on AMI and DIRHA multi-array corpora using the encoder-decoder architecture. Compared with the best single-array results, the proposed framework has achieved relative Word Error Rates (WERs) reduction of 3.7% and 9.7% in the two datasets, respectively, which is better than conventional strategies as well.

cs.CL↗

arXiv2019

Charge Transfer Effect under Odd-Parity Crystalline Electric Field: Divergence of Magnetic Toroidal Fluctuation in $β$-YbAlB$_4$

A novel property of the quantum critical heavy fermion superconductor $β$-YbAlB$_4$ is revealed theoretically. By analyzing the crystalline electronic field (CEF) on the basis of the hybridization picture, odd parity CEF is shown to exist because of sevenfold configuration of B atoms around Yb, which breaks the local inversion symmetry. This allows onsite admixture of 4f and 5d wavefunctions with a pure imaginary coefficient, giving rise to the magnetic toroidal (MT) degree of freedom. By constructing a realistic minimal model for $β$-YbAlB$_4$, we show that onsite 4f-5d Coulomb repulsion drives charge transfer between the 4f and 5d states at Yb, which makes the MT fluctuation as well as the electric dipole fluctuation diverge simultaneously with the critical Yb-valence fluctuation at the quantum critical point of the valence transition.

cond-mat.str-el↗

arXiv2018

Vectorization of hypotheses and speech for faster beam search in encoder decoder-based speech recognition

Attention-based encoder decoder network uses a left-to-right beam search algorithm in the inference step. The current beam search expands hypotheses and traverses the expanded hypotheses at the next time step. This traversal is implemented using a for-loop program in general, and it leads to speed down of the recognition process. In this paper, we propose a parallelism technique for beam search, which accelerates the search process by vectorizing multiple hypotheses to eliminate the for-loop program. We also propose a technique to batch multiple speech utterances for off-line recognition use, which reduces the for-loop program with regard to the traverse of multiple utterances. This extension is not trivial during beam search unlike during training due to several pruning and thresholding techniques for efficient decoding. In addition, our method can combine scores of external modules, RNNLM and CTC, in a batch as shallow fusion. We achieved 3.7 x speedup compared with the original beam search algorithm by vectoring hypotheses, and achieved 10.5 x speedup by further changing processing unit to GPU.

cs.SD↗

arXiv2018

Multi-encoder multi-resolution framework for end-to-end speech recognition

Attention-based methods and Connectionist Temporal Classification (CTC) network have been promising research directions for end-to-end Automatic Speech Recognition (ASR). The joint CTC/Attention model has achieved great success by utilizing both architectures during multi-task training and joint decoding. In this work, we present a novel Multi-Encoder Multi-Resolution (MEMR) framework based on the joint CTC/Attention model. Two heterogeneous encoders with different architectures, temporal resolutions and separate CTC networks work in parallel to extract complimentary acoustic information. A hierarchical attention mechanism is then used to combine the encoder-level information. To demonstrate the effectiveness of the proposed model, experiments are conducted on Wall Street Journal (WSJ) and CHiME-4, resulting in relative Word Error Rate (WER) reduction of 18.0-32.1%. Moreover, the proposed MEMR model achieves 3.6% WER in the WSJ eval92 test set, which is the best WER reported for an end-to-end system on this benchmark.

cs.CL↗

arXiv2018

Joint Acoustic and Class Inference for Weakly Supervised Sound Event Detection

Sound event detection is a challenging task, especially for scenes with multiple simultaneous events. While event classification methods tend to be fairly accurate, event localization presents additional challenges, especially when large amounts of labeled data are not available. Task4 of the 2018 DCASE challenge presents an event detection task that requires accuracy in both segmentation and recognition of events while providing only weakly labeled training data. Supervised methods can produce accurate event labels but are limited in event segmentation when training data lacks event timestamps. On the other hand, unsupervised methods that model the acoustic properties of the audio can produce accurate event boundaries but are not guided by the characteristics of event classes and sound categories. We present a hybrid approach that combines an acoustic-driven event boundary detection and a supervised label inference using a deep neural network. This framework leverages benefits of both unsupervised and supervised methodologies and takes advantage of large amounts of unlabeled data, making it ideal for large-scale weakly labeled event detection. Compared to a baseline system, the proposed approach delivers a 15% absolute improvement in F-score, demonstrating the benefits of the hybrid bottom-up, top-down approach.

eess.AS↗

arXiv2018

Promising Accurate Prefix Boosting for sequence-to-sequence ASR

In this paper, we present promising accurate prefix boosting (PAPB), a discriminative training technique for attention based sequence-to-sequence (seq2seq) ASR. PAPB is devised to unify the training and testing scheme in an effective manner. The training procedure involves maximizing the score of each partial correct sequence obtained during beam search compared to other hypotheses. The training objective also includes minimization of token (character) error rate. PAPB shows its efficacy by achieving 10.8\% and 3.8\% WER with and without RNNLM respectively on Wall Street Journal dataset.

eess.AS↗

arXiv2018

Analysis of Multilingual Sequence-to-Sequence speech recognition systems

This paper investigates the applications of various multilingual approaches developed in conventional hidden Markov model (HMM) systems to sequence-to-sequence (seq2seq) automatic speech recognition (ASR). On a set composed of Babel data, we first show the effectiveness of multi-lingual training with stacked bottle-neck (SBN) features. Then we explore various architectures and training strategies of multi-lingual seq2seq models based on CTC-attention networks including combinations of output layer, CTC and/or attention component re-training. We also investigate the effectiveness of language-transfer learning in a very low resource scenario when the target language is not included in the original multi-lingual training data. Interestingly, we found multilingual features superior to multilingual models, and this finding suggests that we can efficiently combine the benefits of the HMM system with the seq2seq system through these multilingual feature techniques.

eess.AS↗

arXiv2018

Building Corpora for Single-Channel Speech Separation Across Multiple Domains

To date, the bulk of research on single-channel speech separation has been conducted using clean, near-field, read speech, which is not representative of many modern applications. In this work, we develop a procedure for constructing high-quality synthetic overlap datasets, necessary for most deep learning-based separation frameworks. We produced datasets that are more representative of realistic applications using the CHiME-5 and Mixer 6 corpora and evaluate standard methods on this data to demonstrate the shortcomings of current source-separation performance. We also demonstrate the value of a wide variety of data in training robust models that generalize well to multiple conditions.

cs.CL↗

arXiv2018

End-to-End Monaural Multi-speaker ASR System without Pretraining

Recently, end-to-end models have become a popular approach as an alternative to traditional hybrid models in automatic speech recognition (ASR). The multi-speaker speech separation and recognition task is a central task in cocktail party problem. In this paper, we present a state-of-the-art monaural multi-speaker end-to-end automatic speech recognition model. In contrast to previous studies on the monaural multi-speaker speech recognition, this end-to-end framework is trained to recognize multiple label sequences completely from scratch. The system only requires the speech mixture and corresponding label sequences, without needing any indeterminate supervisions obtained from non-mixture speech or corresponding labels/alignments. Moreover, we exploited using the individual attention module for each separated speaker and the scheduled sampling to further improve the performance. Finally, we evaluate the proposed model on the 2-speaker mixed speech generated from the WSJ corpus and the wsj0-2mix dataset, which is a speech separation and recognition benchmark. The experiments demonstrate that the proposed methods can improve the performance of the end-to-end model in separating the overlapping speech and recognizing the separated streams. From the results, the proposed model leads to ~10.0% relative performance gains in terms of CER and WER respectively.

cs.CL↗

arXiv2018

Multilingual sequence-to-sequence speech recognition: architecture, transfer learning, and language modeling

Sequence-to-sequence (seq2seq) approach for low-resource ASR is a relatively new direction in speech research. The approach benefits by performing model training without using lexicon and alignments. However, this poses a new problem of requiring more data compared to conventional DNN-HMM systems. In this work, we attempt to use data from 10 BABEL languages to build a multi-lingual seq2seq model as a prior model, and then port them towards 4 other BABEL languages using transfer learning approach. We also explore different architectures for improving the prior multilingual seq2seq model. The paper also discusses the effect of integrating a recurrent neural network language model (RNNLM) with a seq2seq model during decoding. Experimental results show that the transfer learning approach from the multilingual model shows substantial gains over monolingual models across all 4 BABEL languages. Incorporating an RNNLM also brings significant improvements in terms of %WER, and achieves recognition performance comparable to the models trained with twice more training data.

cs.CL↗

arXiv2018

Low-Resource Contextual Topic Identification on Speech

In topic identification (topic ID) on real-world unstructured audio, an audio instance of variable topic shifts is first broken into sequential segments, and each segment is independently classified. We first present a general purpose method for topic ID on spoken segments in low-resource languages, using a cascade of universal acoustic modeling, translation lexicons to English, and English-language topic classification. Next, instead of classifying each segment independently, we demonstrate that exploring the contextual dependencies across sequential segments can provide large improvements. In particular, we propose an attention-based contextual model which is able to leverage the contexts in a selective manner. We test both our contextual and non-contextual models on four LORELEI languages, and on all but one our attention-based contextual model significantly outperforms the context-independent models.

cs.CL↗

arXiv2018

The Dominant Role of Critical Valence Fluctuations on High $T_{\rm c}$ Superconductivity in Heavy Fermions

Despite almost 40 years of research, the origin of heavy-fermion superconductivity is still strongly debated. Especially, the pressure-induced enhancement of superconductivity in CeCu$_2$Si$_2$ away from the magnetic breakdown is not sufficiently taken into consideration. As recently reported in CeCu$_2$Si$_2$ and several related compounds, optimal superconductivity occurs at the pressure of a valence crossover, which arises from a virtual critical end point at negative temperature $T_{\rm cr}$. In this context, we did a meticulous analysis of a vast set of top-quality high-pressure electrical resistivity data of several Ce-based heavy fermion compounds. The key novelty is the salient correlation between the superconducting transition temperature $T_{\rm c}$ and the valence instability parameter $T_{\rm cr}$, which is in line with theory of enhanced valence fluctuations. Moreover, it is found that, in the pressure region of superconductivity, electrical resistivity is governed by the valence crossover, which most often manifests in scaling behavior. We develop the new idea that the optimum superconducting $T_{\rm c}$ of a given sample is mainly controlled by the compound's $T_{\rm cr}$ and limited by non-magnetic disorder. In this regard, the present study provides compelling evidence for the crucial role of critical valence fluctuations in the formation of Cooper pairs in Ce-based heavy fermion superconductors besides the contribution of spin fluctuations near magnetic quantum critical points, and corroborates a plausible superconducting mechanism in strongly correlated electron systems in general.

cond-mat.str-el↗

arXiv2018

On anomalous temperature dependence of relaxation rate measured by \muSR in α-YbAl_{0.986}Fe_{0.014}B_4

Recently, it was reported by MacLaughlin et al. in Phys. Rev. B 93, 214421 (2016) that α-YbAl_{0.986}Fe_{0.014}B_4 exhibits an anomalous temperature dependence in the relaxation rate 1/T_{1} of $μ$SR, and stressed that such temperature dependence cannot be understood by the scenario based on the quantum critical valence transition (QCVT) while this compound exhibits a series of the non-Fermi liquid behaviors explained by the theory of the QCVT. In this paper, we point out that the anomalous temperature dependence in 1/T_{1} can be understood semi-quantitatively by assuming that the attraction of a screening cloud of conduction electrons about the μ^{+} induces a local magnetic moment arising from a 4f hole on the Yb ion, giving rise to the Kondo effect between heavy quasiparticles.

cond-mat.str-el↗

arXiv2018

End-to-end Speech Recognition with Word-based RNN Language Models

This paper investigates the impact of word-based RNN language models (RNN-LMs) on the performance of end-to-end automatic speech recognition (ASR). In our prior work, we have proposed a multi-level LM, in which character-based and word-based RNN-LMs are combined in hybrid CTC/attention-based ASR. Although this multi-level approach achieves significant error reduction in the Wall Street Journal (WSJ) task, two different LMs need to be trained and used for decoding, which increase the computational cost and memory usage. In this paper, we further propose a novel word-based RNN-LM, which allows us to decode with only the word-based LM, where it provides look-ahead word probabilities to predict next characters instead of the character-based LM, leading competitive accuracy with less computation compared to the multi-level LM. We demonstrate the efficacy of the word-based RNN-LMs using a larger corpus, LibriSpeech, in addition to WSJ we used in the prior work. Furthermore, we show that the proposed model achieves 5.1 %WER for WSJ Eval'92 test set when the vocabulary size is increased, which is the best WER reported for end-to-end ASR systems on this benchmark.

cs.CL↗

arXiv2018

Multi-Head Decoder for End-to-End Speech Recognition

This paper presents a new network architecture called multi-head decoder for end-to-end speech recognition as an extension of a multi-head attention model. In the multi-head attention model, multiple attentions are calculated, and then, they are integrated into a single attention. On the other hand, instead of the integration in the attention level, our proposed method uses multiple decoders for each attention and integrates their outputs to generate a final output. Furthermore, in order to make each head to capture the different modalities, different attention functions are used for each head, leading to the improvement of the recognition performance with an ensemble effect. To evaluate the effectiveness of our proposed method, we conduct an experimental evaluation using Corpus of Spontaneous Japanese. Experimental results demonstrate that our proposed method outperforms the conventional methods such as location-based and multi-head attention models, and that it can capture different speech/linguistic contexts within the attention-based encoder-decoder framework.

cs.CL↗

arXiv2018

Back-Translation-Style Data Augmentation for End-to-End ASR

In this paper we propose a novel data augmentation method for attention-based end-to-end automatic speech recognition (E2E-ASR), utilizing a large amount of text which is not paired with speech signals. Inspired by the back-translation technique proposed in the field of machine translation, we build a neural text-to-encoder model which predicts a sequence of hidden states extracted by a pre-trained E2E-ASR encoder from a sequence of characters. By using hidden states as a target instead of acoustic features, it is possible to achieve faster attention learning and reduce computational cost, thanks to sub-sampling in E2E-ASR encoder, also the use of the hidden states can avoid to model speaker dependencies unlike acoustic features. After training, the text-to-encoder model generates the hidden states from a large amount of unpaired text, then E2E-ASR decoder is retrained using the generated hidden states as additional training data. Experimental evaluation using LibriSpeech dataset demonstrates that our proposed method achieves improvement of ASR performance and reduces the number of unknown words without the need for paired data.

cs.CL↗

arXiv2018

Multi-Modal Data Augmentation for End-to-End ASR

We present a new end-to-end architecture for automatic speech recognition (ASR) that can be trained using \emph{symbolic} input in addition to the traditional acoustic input. This architecture utilizes two separate encoders: one for acoustic input and another for symbolic input, both sharing the attention and decoder parameters. We call this architecture a multi-modal data augmentation network (MMDA), as it can support multi-modal (acoustic and symbolic) input and enables seamless mixing of large text datasets with significantly smaller transcribed speech corpora during training. We study different ways of transforming large text corpora into a symbolic form suitable for training our MMDA network. Our best MMDA setup obtains small improvements on character error rate (CER), and as much as 7-10\% relative word error rate (WER) improvement over a baseline both with and without an external language model.

cs.CL↗

arXiv2018

A Purely End-to-end System for Multi-speaker Speech Recognition

Recently, there has been growing interest in multi-speaker speech recognition, where the utterances of multiple speakers are recognized from their mixture. Promising techniques have been proposed for this task, but earlier works have required additional training data such as isolated source signals or senone alignments for effective learning. In this paper, we propose a new sequence-to-sequence framework to directly decode multiple label sequences from a single speech sequence by unifying source separation and speech recognition functions in an end-to-end manner. We further propose a new objective function to improve the contrast between the hidden vectors to avoid generating similar hypotheses. Experimental results show that the model is directly able to learn a mapping from a speech mixture to multiple label sequences, achieving 83.1 % relative improvement compared to a model trained without the proposed objective. Interestingly, the results are comparable to those produced by previous end-to-end works featuring explicit separation and recognition modules.

cs.SD↗