Li-Rong Dai

Ziqiang Zhang, Sanyuan Chen, Long Zhou et al. · microsoft-research

How to boost speech pre-training with textual data is an unsolved problem due to the fact that speech and text are very different modalities with distinct characteristics. In this paper, we propose a cross-modal Speech and Language Model (SpeechLM) to explicitly align speech and text pre-training with a pre-defined unified discrete representation. Specifically, we introduce two alternative discrete tokenizers to bridge the speech and text modalities, including phoneme-unit and hidden-unit tokenizers, which can be trained using a small amount of paired speech-text data. Based on the trained tokenizers, we convert the unlabeled speech and text data into tokens of phoneme units or hidden units. The pre-training objective is designed to unify the speech and the text into the same discrete semantic space with a unified Transformer network. We evaluate SpeechLM on various spoken language processing tasks including speech recognition, speech translation, and universal representation evaluation framework SUPERB, demonstrating significant improvements on content-related tasks. Code and models are available at https://aka.ms/SpeechLM.

Ziqiang Zhang, Long Zhou, Junyi Ao et al. · microsoft-research

The rapid development of single-modal pre-training has prompted researchers to pay more attention to cross-modal pre-training methods. In this paper, we propose a unified-modal speech-unit-text pre-training model, SpeechUT, to connect the representations of a speech encoder and a text decoder with a shared unit encoder. Leveraging hidden-unit as an interface to align speech and text, we can decompose the speech-to-text model into a speech-to-unit model and a unit-to-text model, which can be jointly pre-trained with unpaired speech and text data respectively. Our proposed SpeechUT is fine-tuned and evaluated on automatic speech recognition (ASR) and speech translation (ST) tasks. Experimental results show that SpeechUT gets substantial improvements over strong baselines, and achieves state-of-the-art performance on both the LibriSpeech ASR and MuST-C ST tasks. To better understand the proposed SpeechUT, detailed analyses are conducted. The code and pre-trained models are available at https://aka.ms/SpeechUT.

Qiushi Zhu, Long Zhou, Ziqiang Zhang et al. · microsoft-research

Although speech is a simple and effective way for humans to communicate with the outside world, a more realistic speech interaction contains multimodal information, e.g., vision, text. How to design a unified framework to integrate different modal information and leverage different resources (e.g., visual-audio pairs, audio-text pairs, unlabeled speech, and unlabeled text) to facilitate speech representation learning was not well explored. In this paper, we propose a unified cross-modal representation learning framework VATLM (Visual-Audio-Text Language Model). The proposed VATLM employs a unified backbone network to model the modality-independent information and utilizes three simple modality-dependent modules to preprocess visual, speech, and text inputs. In order to integrate these three modalities into one shared semantic space, VATLM is optimized with a masked prediction task of unified tokens, given by our proposed unified tokenizer. We evaluate the pre-trained VATLM on audio-visual related downstream tasks, including audio-visual speech recognition (AVSR), visual speech recognition (VSR) tasks. Results show that the proposed VATLM outperforms previous the state-of-the-art models, such as audio-visual pre-trained AV-HuBERT model, and analysis also demonstrates that VATLM is capable of aligning different modalities into the same space. To facilitate future research, we release the code and pre-trained models at https://aka.ms/vatlm.

Ye-Qian Du, Jie Zhang, Qiu-Shi Zhu et al.

Unpaired data has shown to be beneficial for low-resource automatic speech recognition~(ASR), which can be involved in the design of hybrid models with multi-task training or language model dependent pre-training. In this work, we leverage unpaired data to train a general sequence-to-sequence model. Unpaired speech and text are used in the form of data pairs by generating the corresponding missing parts in prior to model training. Inspired by the complementarity of speech-PseudoLabel pair and SynthesizedAudio-text pair in both acoustic features and linguistic features, we propose a complementary joint training~(CJT) method that trains a model alternatively with two data pairs. Furthermore, label masking for pseudo-labels and gradient restriction for synthesized audio are proposed to further cope with the deviations from real data, termed as CJT++. Experimental results show that compared to speech-only training, the proposed basic CJT achieves great performance improvements on clean/other test sets, and the CJT++ re-training yields further performance enhancements. It is also apparent that the proposed method outperforms the wav2vec2.0 model with the same model size and beam size, particularly in extreme low-resource cases.

Jian Zhu, Yu Cui, Zhangmin Huang et al.

The multi-view hash method converts heterogeneous data from multiple views into binary hash codes, which is one of the critical technologies in multimedia retrieval. However, the current methods mainly explore the complementarity among multiple views while lacking confidence learning and fusion. Moreover, in practical application scenarios, the single-view data contain redundant noise. To conduct the confidence learning and eliminate unnecessary noise, we propose a novel Adaptive Confidence Multi-View Hashing (ACMVH) method. First, a confidence network is developed to extract useful information from various single-view features and remove noise information. Furthermore, an adaptive confidence multi-view network is employed to measure the confidence of each view and then fuse multi-view features through a weighted summation. Lastly, a dilation network is designed to further enhance the feature representation of the fused features. To the best of our knowledge, we pioneer the application of confidence learning into the field of multimedia retrieval. Extensive experiments on two public datasets show that the proposed ACMVH performs better than state-of-the-art methods (maximum increase of 3.24%). The source code is available at https://github.com/HackerHyper/ACMVH.

Mengzhi Wang, Shifu Xiong, Genshun Wan et al.

Contextual-LAS (CLAS) has been shown effective in improving Automatic Speech Recognition (ASR) of rare words. It relies on phrase-level contextual modeling and attention-based relevance scoring without explicit contextual constraint which lead to insufficient use of contextual information. In this work, we propose deep CLAS to use contextual information better. We introduce bias loss forcing model to focus on contextual information. The query of bias attention is also enriched to improve the accuracy of the bias attention score. To get fine-grained contextual information, we replace phrase-level encoding with character-level encoding and encode contextual information with conformer rather than LSTM. Moreover, we directly use the bias attention score to correct the output probability distribution of the model. Experiments using the public AISHELL-1 and AISHELL-NER. On AISHELL-1, compared to CLAS baselines, deep CLAS obtains a 65.78% relative recall and a 53.49% relative F1-score increase in the named entity recognition scene.

Jian Zhu, Xin Zou, Lei Liu et al.

Multi-view clustering can partition data samples into their categories by learning a consensus representation in an unsupervised way and has received more and more attention in recent years. However, there is an untrusted fusion problem. The reasons for this problem are as follows: 1) The current methods ignore the presence of noise or redundant information in the view; 2) The similarity of contrastive learning comes from the same sample rather than the same cluster in deep multi-view clustering. It causes multi-view fusion in the wrong direction. This paper proposes a novel multi-view clustering network to address this problem, termed as Trusted Mamba Contrastive Network (TMCN). Specifically, we present a new Trusted Mamba Fusion Network (TMFN), which achieves a trusted fusion of multi-view data through a selective mechanism. Moreover, we align the fused representation and the view-specific representation using the Average-similarity Contrastive Learning (AsCL) module. AsCL increases the similarity of view presentation from the same cluster, not merely from the same sample. Extensive experiments show that the proposed method achieves state-of-the-art results in deep multi-view clustering tasks. The source code is available at https://github.com/HackerHyper/TMCN.

Junyi Ao, Ziqiang Zhang, Long Zhou et al.

This paper studies a novel pre-training technique with unpaired speech data, Speech2C, for encoder-decoder based automatic speech recognition (ASR). Within a multi-task learning framework, we introduce two pre-training tasks for the encoder-decoder network using acoustic units, i.e., pseudo codes, derived from an offline clustering model. One is to predict the pseudo codes via masked language modeling in encoder output, like HuBERT model, while the other lets the decoder learn to reconstruct pseudo codes autoregressively instead of generating textual scripts. In this way, the decoder learns to reconstruct original speech information with codes before learning to generate correct text. Comprehensive experiments on the LibriSpeech corpus show that the proposed Speech2C can relatively reduce the word error rate (WER) by 19.2% over the method without decoder pre-training, and also outperforms significantly the state-of-the-art wav2vec 2.0 and HuBERT on fine-tuning subsets of 10h and 100h. We release our code and model at https://github.com/microsoft/SpeechT5/tree/main/Speech2C.

Jianwei Cui, Yu Gu, Chao Weng et al.

This paper presents an advanced end-to-end singing voice synthesis (SVS) system based on the source-filter mechanism that directly translates lyrical and melodic cues into expressive and high-fidelity human-like singing. Similarly to VISinger 2, the proposed system also utilizes training paradigms evolved from VITS and incorporates elements like the fundamental pitch (F0) predictor and waveform generation decoder. To address the issue that the coupling of mel-spectrogram features with F0 information may introduce errors during F0 prediction, we consider two strategies. Firstly, we leverage mel-cepstrum (mcep) features to decouple the intertwined mel-spectrogram and F0 characteristics. Secondly, inspired by the neural source-filter models, we introduce source excitation signals as the representation of F0 in the SVS system, aiming to capture pitch nuances more accurately. Meanwhile, differentiable mcep and F0 losses are employed as the waveform decoder supervision to fortify the prediction accuracy of speech envelope and pitch in the generated speech. Experiments on the Opencpop dataset demonstrate efficacy of the proposed model in synthesis quality and intonation accuracy.

Zi-Qiang Zhang, Jie Zhang, Jian-Shu Zhang et al.

With the advance in self-supervised learning for audio and visual modalities, it has become possible to learn a robust audio-visual speech representation. This would be beneficial for improving the audio-visual speech recognition (AVSR) performance, as the multi-modal inputs contain more fruitful information in principle. In this paper, based on existing self-supervised representation learning methods for audio modality, we therefore propose an audio-visual representation learning approach. The proposed approach explores both the complementarity of audio-visual modalities and long-term context dependency using a transformer-based fusion module and a flexible masking strategy. After pre-training, the model is able to extract fused representations required by AVSR. Without loss of generality, it can be applied to single-modal tasks, e.g. audio/visual speech recognition by simply masking out one modality in the fusion module. The proposed pre-trained model is evaluated on speech recognition and lipreading tasks using one or two modalities, where the superiority is revealed.

Xing-Yu Chen, Qiu-Shi Zhu, Jie Zhang et al.

In this work, we propose a bi-directional long short-term memory (BiLSTM) network based COVID-19 detection method using breath/speech/cough signals. By using the acoustic signals to train the network, respectively, we can build individual models for three tasks, whose parameters are averaged to obtain an average model, which is then used as the initialization for the BiLSTM model training of each task. This initialization method can significantly improve the performance on the three tasks, which surpasses the official baseline results. Besides, we also utilize a public pre-trained model wav2vec2.0 and pre-train it using the official DiCOVA datasets. This wav2vec2.0 model is utilized to extract high-level features of the sound as the model input to replace conventional mel-frequency cepstral coefficients (MFCC) features. Experimental results reveal that using high-level features together with MFCC features can improve the performance. To further improve the performance, we also deploy some preprocessing techniques like silent segment removal, amplitude normalization and time-frequency mask. The proposed detection model is evaluated on the DiCOVA dataset and results show that our method achieves an area under curve (AUC) score of 88.44% on blind test in the fusion track.

Qiu-Shi Zhu, Jie Zhang, Zi-Qiang Zhang et al.

Wav2vec2.0 is a popular self-supervised pre-training framework for learning speech representations in the context of automatic speech recognition (ASR). It was shown that wav2vec2.0 has a good robustness against the domain shift, while the noise robustness is still unclear. In this work, we therefore first analyze the noise robustness of wav2vec2.0 via experiments. We observe that wav2vec2.0 pre-trained on noisy data can obtain good representations and thus improve the ASR performance on the noisy test set, which however brings a performance degradation on the clean test set. To avoid this issue, in this work we propose an enhanced wav2vec2.0 model. Specifically, the noisy speech and the corresponding clean version are fed into the same feature encoder, where the clean speech provides training targets for the model. Experimental results reveal that the proposed method can not only improve the ASR performance on the noisy test set which surpasses the original wav2vec2.0, but also ensure a tiny performance decrease on the clean test set. In addition, the effectiveness of the proposed method is demonstrated under different types of noise conditions.

Dan Liu, Mengge Du, Xiaoxi Li et al.

This paper describes USTC-NELSLIP's submissions to the IWSLT2021 Simultaneous Speech Translation task. We proposed a novel simultaneous translation model, Cross Attention Augmented Transducer (CAAT), which extends conventional RNN-T to sequence-to-sequence tasks without monotonic constraints, e.g., simultaneous translation. Experiments on speech-to-text (S2T) and text-to-text (T2T) simultaneous translation tasks shows CAAT achieves better quality-latency trade-offs compared to \textit{wait-k}, one of the previous state-of-the-art approaches. Based on CAAT architecture and data augmentation, we build S2T and T2T simultaneous translation systems in this evaluation campaign. Compared to last year's optimal systems, our S2T simultaneous translation system improves by an average of 11.3 BLEU for all latency regimes, and our T2T simultaneous translation system improves by an average of 4.6 BLEU.

Zi-Qiang Zhang, Yan Song, Ming-Hui Wu et al.

In this paper, we propose a weakly supervised multilingual representation learning framework, called cross-lingual self-training (XLST). XLST is able to utilize a small amount of annotated data from high-resource languages to improve the representation learning on multilingual un-annotated data. Specifically, XLST uses a supervised trained model to produce initial representations and another model to learn from them, by maximizing the similarity between output embeddings of these two models. Furthermore, the moving average mechanism and multi-view data augmentation are employed, which are experimentally shown to be crucial to XLST. Comprehensive experiments have been conducted on the CommonVoice corpus to evaluate the effectiveness of XLST. Results on 5 downstream low-resource ASR tasks shows that our multilingual pretrained model achieves relatively 18.6% PER reduction over the state-of-the-art self-supervised method, with leveraging additional 100 hours of annotated English data.

Hang Chen, Jun Du, Yu Hu et al.

In this paper, we propose a novel deep learning architecture to improving word-level lip-reading. On the one hand, we first introduce the multi-scale processing into the spatial feature extraction for lip-reading. Specially, we proposed hierarchical pyramidal convolution (HPConv) to replace the standard convolution in original module, leading to improvements over the model's ability to discover fine-grained lip movements. On the other hand, we merge information in all time steps of the sequence by utilizing self-attention, to make the model pay more attention to the relevant frames. These two advantages are combined together to further enhance the model's classification power. Experiments on the Lip Reading in the Wild (LRW) dataset show that our proposed model has achieved 86.83% accuracy, yielding 1.53% absolute improvement over the current state-of-the-art. We also conducted extensive experiments to better understand the behavior of the proposed model.

Hang Chen, Jun Du, Yu Hu et al.

In this paper, we propose a visual embedding approach to improving embedding aware speech enhancement (EASE) by synchronizing visual lip frames at the phone and place of articulation levels. We first extract visual embedding from lip frames using a pre-trained phone or articulation place recognizer for visual-only EASE (VEASE). Next, we extract audio-visual embedding from noisy speech and lip videos in an information intersection manner, utilizing a complementarity of audio and visual features for multi-modal EASE (MEASE). Experiments on the TCD-TIMIT corpus corrupted by simulated additive noises show that our proposed subword based VEASE approach is more effective than conventional embedding at the word level. Moreover, visual embedding at the articulation place level, leveraging upon a high correlation between place of articulation and lip shapes, shows an even better performance than that at the phone level. Finally the proposed MEASE framework, incorporating both audio and visual embedding, yields significantly better speech quality and intelligibility than those obtained with the best visual-only and audio-only EASE systems.

Jing-Xuan Zhang, Li-Juan Liu, Yan-Nian Chen et al.

With the development of automatic speech recognition (ASR) and text-to-speech synthesis (TTS) technique, it's intuitive to construct a voice conversion system by cascading an ASR and TTS system. In this paper, we present a ASR-TTS method for voice conversion, which used iFLYTEK ASR engine to transcribe the source speech into text and a Transformer TTS model with WaveNet vocoder to synthesize the converted speech from the decoded text. For the TTS model, we proposed to use a prosody code to describe the prosody information other than text and speaker information contained in speech. A prosody encoder is used to extract the prosody code. During conversion, the source prosody is transferred to converted speech by conditioning the Transformer TTS model with its code. Experiments were conducted to demonstrate the effectiveness of our proposed method. Our system also obtained the best naturalness and similarity in the mono-lingual task of Voice Conversion Challenge 2020.

Liangfa Wei, Jie Zhang, Junfeng Hou et al.

Audio-visual information fusion enables a performance improvement in speech recognition performed in complex acoustic scenarios, e.g., noisy environments. It is required to explore an effective audio-visual fusion strategy for audiovisual alignment and modality reliability. Different from the previous end-to-end approaches where the audio-visual fusion is performed after encoding each modality, in this paper we propose to integrate an attentive fusion block into the encoding process. It is shown that the proposed audio-visual fusion method in the encoder module can enrich audio-visual representations, as the relevance between the two modalities is leveraged. In line with the transformer-based architecture, we implement the embedded fusion block using a multi-head attention based audiovisual fusion with one-way or two-way interactions. The proposed method can sufficiently combine the two streams and weaken the over-reliance on the audio modality. Experiments on the LRS3-TED dataset demonstrate that the proposed method can increase the recognition rate by 0.55%, 4.51% and 4.61% on average under the clean, seen and unseen noise conditions, respectively, compared to the state-of-the-art approach.

Jing-Xuan Zhang, Zhen-Hua Ling, Li-Rong Dai

This paper presents an adversarial learning method for recognition-synthesis based non-parallel voice conversion. A recognizer is used to transform acoustic features into linguistic representations while a synthesizer recovers output features from the recognizer outputs together with the speaker identity. By separating the speaker characteristics from the linguistic representations, voice conversion can be achieved by replacing the speaker identity with the target one. In our proposed method, a speaker adversarial loss is adopted in order to obtain speaker-independent linguistic representations using the recognizer. Furthermore, discriminators are introduced and a generative adversarial network (GAN) loss is used to prevent the predicted features from being over-smoothed. For training model parameters, a strategy of pre-training on a multi-speaker dataset and then fine-tuning on the source-target speaker pair is designed. Our method achieved higher similarity than the baseline model that obtained the best performance in Voice Conversion Challenge 2018.

Fenglin Ding, Wu Guo, Lirong Dai et al.

Batch normalization (BN) is an effective method to accelerate model training and improve the generalization performance of neural networks. In this paper, we propose an improved batch normalization technique called attentive batch normalization (ABN) in Long Short Term Memory (LSTM) based acoustic modeling for automatic speech recognition (ASR). In the proposed method, an auxiliary network is used to dynamically generate the scaling and shifting parameters in batch normalization, and attention mechanisms are introduced to improve their regularized performance. Furthermore, two schemes, frame-level and utterance-level ABN, are investigated. We evaluate our proposed methods on Mandarin and Uyghur ASR tasks, respectively. The experimental results show that the proposed ABN greatly improves the performance of batch normalization in terms of transcription accuracy for both languages.

Peng-fei Wu, Zhen-hua Ling, Li-juan Liu et al.

This paper proposes an end-to-end emotional speech synthesis (ESS) method which adopts global style tokens (GSTs) for semi-supervised training. This model is built based on the GST-Tacotron framework. The style tokens are defined to present emotion categories. A cross entropy loss function between token weights and emotion labels is designed to obtain the interpretability of style tokens utilizing the small portion of training data with emotion labels. Emotion recognition experiments confirm that this method can achieve one-to-one correspondence between style tokens and emotion categories effectively. Objective and subjective evaluation results show that our model outperforms the conventional Tacotron model for ESS when only 5\% of training data has emotion labels. Its subjective performance is close to the Tacotron model trained using all emotion labels.

Yuan-Hao Yi, Yang Ai, Zhen-Hua Ling et al.

This paper presents a method of using autoregressive neural networks for the acoustic modeling of singing voice synthesis (SVS). Singing voice differs from speech and it contains more local dynamic movements of acoustic features, e.g., vibratos. Therefore, our method adopts deep autoregressive (DAR) models to predict the F0 and spectral features of singing voice in order to better describe the dependencies among the acoustic features of consecutive frames. For F0 modeling, discretized F0 values are used and the influences of the history length in DAR are analyzed by experiments. An F0 post-processing strategy is also designed to alleviate the inconsistency between the predicted F0 contours and the F0 values determined by music notes. Furthermore, we extend the DAR model to deal with continuous spectral features, and a prenet module with self-attention layers is introduced to process historical frames. Experiments on a Chinese singing voice corpus demonstrate that our method using DARs can produce F0 contours with vibratos effectively, and can achieve better objective and subjective performance than the conventional method using recurrent neural networks (RNNs).

Lanhua You, Wu Guo, Lirong Dai et al.

In this paper, gating mechanisms are applied in deep neural network (DNN) training for x-vector-based text-independent speaker verification. First, a gated convolution neural network (GCNN) is employed for modeling the frame-level embedding layers. Compared with the time-delay DNN (TDNN), the GCNN can obtain more expressive frame-level representations through carefully designed memory cell and gating mechanisms. Moreover, we propose a novel gated-attention statistics pooling strategy in which the attention scores are shared with the output gate. The gated-attention statistics pooling combines both gating and attention mechanisms into one framework; therefore, we can capture more useful information in the temporal pooling layer. Experiments are carried out using the NIST SRE16 and SRE18 evaluation datasets. The results demonstrate the effectiveness of the GCNN and show that the proposed gated-attention statistics pooling can further improve the performance.

Lanhua You, Wu Guo, Lirong Dai et al.

The x-vector based deep neural network (DNN) embedding systems have demonstrated effectiveness for text-independent speaker verification. This paper presents a multi-task learning architecture for training the speaker embedding DNN with the primary task of classifying the target speakers, and the auxiliary task of reconstructing the first- and higher-order statistics of the original input utterance. The proposed training strategy aggregates both the supervised and unsupervised learning into one framework to make the speaker embeddings more discriminative and robust. Experiments are carried out using the NIST SRE16 evaluation dataset and the VOiCES dataset. The results demonstrate that our proposed method outperforms the original x-vector approach with very low additional complexity added.

Jing-Xuan Zhang, Zhen-Hua Ling, Yuan Jiang et al.

This paper presents methods of making using of text supervision to improve the performance of sequence-to-sequence (seq2seq) voice conversion. Compared with conventional frame-to-frame voice conversion approaches, the seq2seq acoustic modeling method proposed in our previous work achieved higher naturalness and similarity. In this paper, we further improve its performance by utilizing the text transcriptions of parallel training data. First, a multi-task learning structure is designed which adds auxiliary classifiers to the middle layers of the seq2seq model and predicts linguistic labels as a secondary task. Second, a data-augmentation method is proposed which utilizes text alignment to produce extra parallel sequences for model training. Experiments are conducted to evaluate our proposed method with training sets at different sizes. Experimental results show that the multi-task learning with linguistic labels is effective at reducing the errors of seq2seq voice conversion. The data-augmentation method can further improve the performance of seq2seq voice conversion when only 50 or 100 training utterances are available.

Yaming Liu, Jian Tang, Yan Song et al.

Polyphonic sound event detection (polyphonic SED) is an interesting but challenging task due to the concurrence of multiple sound events. Recently, SED methods based on convolutional neural networks (CNN) and recurrent neural networks (RNN) have shown promising performance. Generally, CNN are designed for local feature extraction while RNN are used to model the temporal dependency among these local features. Despite their success, it is still insufficient for existing deep learning techniques to separate individual sound event from their mixture, largely due to the overlapping characteristic of features. Motivated by the success of Capsule Networks (CapsNet), we propose a more suitable capsule based approach for polyphonic SED. Specifically, several capsule layers are designed to effectively select representative frequency bands for each individual sound event. The temporal dependency of capsule's outputs is then modeled by a RNN. And a dynamic threshold method is proposed for making the final decision based on RNN outputs. Experiments on the TUT-SED Synthetic 2016 dataset show that the proposed approach obtains an F1-score of 68.8% and an error rate of 0.45, outperforming the previous state-of-the-art method of 66.4% and 0.48, respectively.

Jing-Xuan Zhang, Zhen-Hua Ling, Li-Rong Dai

This paper proposes a forward attention method for the sequenceto- sequence acoustic modeling of speech synthesis. This method is motivated by the nature of the monotonic alignment from phone sequences to acoustic sequences. Only the alignment paths that satisfy the monotonic condition are taken into consideration at each decoder timestep. The modified attention probabilities at each timestep are computed recursively using a forward algorithm. A transition agent for forward attention is further proposed, which helps the attention mechanism to make decisions whether to move forward or stay at each decoder timestep. Experimental results show that the proposed forward attention method achieves faster convergence speed and higher stability than the baseline attention method. Besides, the method of forward attention with transition agent can also help improve the naturalness of synthetic speech and control the speed of synthetic speech effectively.

Shiliang Zhang, Ming Lei, Zhijie Yan et al.

In this paper, we present an improved feedforward sequential memory networks (FSMN) architecture, namely Deep-FSMN (DFSMN), by introducing skip connections between memory blocks in adjacent layers. These skip connections enable the information flow across different layers and thus alleviate the gradient vanishing problem when building very deep structure. As a result, DFSMN significantly benefits from these skip connections and deep structure. We have compared the performance of DFSMN to BLSTM both with and without lower frame rate (LFR) on several large speech recognition tasks, including English and Mandarin. Experimental results shown that DFSMN can consistently outperform BLSTM with dramatic gain, especially trained with LFR using CD-Phone as modeling units. In the 2000 hours Fisher (FSH) task, the proposed DFSMN can achieve a word error rate of 9.4% by purely using the cross-entropy criterion and decoding with a 3-gram language model, which achieves a 1.5% absolute improvement compared to the BLSTM. In a 20000 hours Mandarin recognition task, the LFR trained DFSMN can achieve more than 20% relative improvement compared to the LFR trained BLSTM. Moreover, we can easily design the lookahead filter order of the memory blocks in DFSMN to control the latency for real-time applications.

Zhen-Hua Ling, Yang Ai, Yu Gu et al.

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 composed of long short-term memory (LSTM) layers and feed-forward (FF) layers. The LSTM layers form a hierarchical structure and each layer operates at a specific temporal resolution to efficiently capture long-span dependencies between temporal sequences. Furthermore, additional conditions, such as the bottleneck (BN) features derived from narrowband speech using a deep neural network (DNN)-based state classifier, are employed as auxiliary input to further improve the quality of generated wideband speech. The experimental results of comparing several waveform modeling methods show that the HRNN-based method can achieve better speech quality and run-time efficiency than the dilated convolutional neural network (DCNN)-based method and the plain sample-level recurrent neural network (SRNN)-based method. Our proposed method also outperforms the conventional vocoder-based BWE method using LSTM-RNNs in terms of the subjective quality of the reconstructed wideband speech.

Jianshu Zhang, Yixing Zhu, Jun Du et al.

Recently, great progress has been made for online handwritten Chinese character recognition due to the emergence of deep learning techniques. However, previous research mostly treated each Chinese character as one class without explicitly considering its inherent structure, namely the radical components with complicated geometry. In this study, we propose a novel trajectory-based radical analysis network (TRAN) to firstly identify radicals and analyze two-dimensional structures among radicals simultaneously, then recognize Chinese characters by generating captions of them based on the analysis of their internal radicals. The proposed TRAN employs recurrent neural networks (RNNs) as both an encoder and a decoder. The RNN encoder makes full use of online information by directly transforming handwriting trajectory into high-level features. The RNN decoder aims at generating the caption by detecting radicals and spatial structures through an attention model. The manner of treating a Chinese character as a two-dimensional composition of radicals can reduce the size of vocabulary and enable TRAN to possess the capability of recognizing unseen Chinese character classes, only if the corresponding radicals have been seen. Evaluated on CASIA-OLHWDB database, the proposed approach significantly outperforms the state-of-the-art whole-character modeling approach with a relative character error rate (CER) reduction of 10%. Meanwhile, for the case of recognition of 500 unseen Chinese characters, TRAN can achieve a character accuracy of about 60% while the traditional whole-character method has no capability to handle them.

Jianshu Zhang, Jun Du, Lirong Dai

Handwritten mathematical expression recognition is a challenging problem due to the complicated two-dimensional structures, ambiguous handwriting input and variant scales of handwritten math symbols. To settle this problem, we utilize the attention based encoder-decoder model that recognizes mathematical expression images from two-dimensional layouts to one-dimensional LaTeX strings. We improve the encoder by employing densely connected convolutional networks as they can strengthen feature extraction and facilitate gradient propagation especially on a small training set. We also present a novel multi-scale attention model which is employed to deal with the recognition of math symbols in different scales and save the fine-grained details that will be dropped by pooling operations. Validated on the CROHME competition task, the proposed method significantly outperforms the state-of-the-art methods with an expression recognition accuracy of 52.8% on CROHME 2014 and 50.1% on CROHME 2016, by only using the official training dataset.

Jianshu Zhang, Jun Du, Lirong Dai

In this study, we present a novel end-to-end approach based on the encoder-decoder framework with the attention mechanism for online handwritten mathematical expression recognition (OHMER). First, the input two-dimensional ink trajectory information of handwritten expression is encoded via the gated recurrent unit based recurrent neural network (GRU-RNN). Then the decoder is also implemented by the GRU-RNN with a coverage-based attention model. The proposed approach can simultaneously accomplish the symbol recognition and structural analysis to output a character sequence in LaTeX format. Validated on the CROHME 2014 competition task, our approach significantly outperforms the state-of-the-art with an expression recognition accuracy of 52.43% by only using the official training dataset. Furthermore, the alignments between the input trajectories of handwritten expressions and the output LaTeX sequences are visualized by the attention mechanism to show the effectiveness of the proposed method.

Jianshu Zhang, Yixing Zhu, Jun Du et al.

Chinese characters have a huge set of character categories, more than 20,000 and the number is still increasing as more and more novel characters continue being created. However, the enormous characters can be decomposed into a compact set of about 500 fundamental and structural radicals. This paper introduces a novel radical analysis network (RAN) to recognize printed Chinese characters by identifying radicals and analyzing two-dimensional spatial structures among them. The proposed RAN first extracts visual features from input by employing convolutional neural networks as an encoder. Then a decoder based on recurrent neural networks is employed, aiming at generating captions of Chinese characters by detecting radicals and two-dimensional structures through a spatial attention mechanism. The manner of treating a Chinese character as a composition of radicals rather than a single character class largely reduces the size of vocabulary and enables RAN to possess the ability of recognizing unseen Chinese character classes, namely zero-shot learning.

Yong Xu, Jun Du, Zhen Huang et al.

We propose a multi-objective framework to learn both secondary targets not directly related to the intended task of speech enhancement (SE) and the primary target of the clean log-power spectra (LPS) features to be used directly for constructing the enhanced speech signals. In deep neural network (DNN) based SE we introduce an auxiliary structure to learn secondary continuous features, such as mel-frequency cepstral coefficients (MFCCs), and categorical information, such as the ideal binary mask (IBM), and integrate it into the original DNN architecture for joint optimization of all the parameters. This joint estimation scheme imposes additional constraints not available in the direct prediction of LPS, and potentially improves the learning of the primary target. Furthermore, the learned secondary information as a byproduct can be used for other purposes, e.g., the IBM-based post-processing in this work. A series of experiments show that joint LPS and MFCC learning improves the SE performance, and IBM-based post-processing further enhances listening quality of the reconstructed speech.

Junbei Zhang, Xiaodan Zhu, Qian Chen et al.

The last several years have seen intensive interest in exploring neural-network-based models for machine comprehension (MC) and question answering (QA). In this paper, we approach the problems by closely modelling questions in a neural network framework. We first introduce syntactic information to help encode questions. We then view and model different types of questions and the information shared among them as an adaptation task and proposed adaptation models for them. On the Stanford Question Answering Dataset (SQuAD), we show that these approaches can help attain better results over a competitive baseline.

Shiliang Zhang, Cong Liu, Hui Jiang et al.

In this paper, we propose a novel neural network structure, namely \emph{feedforward sequential memory networks (FSMN)}, to model long-term dependency in time series without using recurrent feedback. The proposed FSMN is a standard fully-connected feedforward neural network equipped with some learnable memory blocks in its hidden layers. The memory blocks use a tapped-delay line structure to encode the long context information into a fixed-size representation as short-term memory mechanism. We have evaluated the proposed FSMNs in several standard benchmark tasks, including speech recognition and language modelling. Experimental results have shown FSMNs significantly outperform the conventional recurrent neural networks (RNN), including LSTMs, in modeling sequential signals like speech or language. Moreover, FSMNs can be learned much more reliably and faster than RNNs or LSTMs due to the inherent non-recurrent model structure.

ShiLiang Zhang, Hui Jiang, Si Wei et al.

We introduce a new structure for memory neural networks, called feedforward sequential memory networks (FSMN), which can learn long-term dependency without using recurrent feedback. The proposed FSMN is a standard feedforward neural networks equipped with learnable sequential memory blocks in the hidden layers. In this work, we have applied FSMN to several language modeling (LM) tasks. Experimental results have shown that the memory blocks in FSMN can learn effective representations of long history. Experiments have shown that FSMN based language models can significantly outperform not only feedforward neural network (FNN) based LMs but also the popular recurrent neural network (RNN) LMs.

Shiliang Zhang, Hui Jiang, Mingbin Xu et al.

In this paper, we propose the new fixed-size ordinally-forgetting encoding (FOFE) method, which can almost uniquely encode any variable-length sequence of words into a fixed-size representation. FOFE can model the word order in a sequence using a simple ordinally-forgetting mechanism according to the positions of words. In this work, we have applied FOFE to feedforward neural network language models (FNN-LMs). Experimental results have shown that without using any recurrent feedbacks, FOFE based FNN-LMs can significantly outperform not only the standard fixed-input FNN-LMs but also the popular RNN-LMs.