arXiv subjects

Wei Chu

Publications and source records attributed to Wei Chu.

At least 73 records · Page 4Linked to original sources

arXiv2021

Low Resource German ASR with Untranscribed Data Spoken by Non-native Children -- INTERSPEECH 2021 Shared Task SPAPL System

This paper describes the SPAPL system for the INTERSPEECH 2021 Challenge: Shared Task on Automatic Speech Recognition for Non-Native Children's Speech in German. ~ 5 hours of transcribed data and ~ 60 hours of untranscribed data are provided to develop a German ASR system for children. For the training of the transcribed data, we propose a non-speech state discriminative loss (NSDL) to mitigate the influence of long-duration non-speech segments within speech utterances. In order to explore the use of the untranscribed data, various approaches are implemented and combined together to incrementally improve the system performance. First, bidirectional autoregressive predictive coding (Bi-APC) is used to learn initial parameters for acoustic modelling using the provided untranscribed data. Second, incremental semi-supervised learning is further used to iteratively generate pseudo-transcribed data. Third, different data augmentation schemes are used at different training stages to increase the variability and size of the training data. Finally, a recurrent neural network language model (RNNLM) is used for rescoring. Our system achieves a word error rate (WER) of 39.68% on the evaluation data, an approximately 12% relative improvement over the official baseline (45.21%).

eess.AS↗

arXiv2021

Symmetric Regularization based BERT for Pair-wise Semantic Reasoning

The ability of semantic reasoning over the sentence pair is essential for many natural language understanding tasks, e.g., natural language inference and machine reading comprehension. A recent significant improvement in these tasks comes from BERT. As reported, the next sentence prediction (NSP) in BERT, which learns the contextual relationship between two sentences, is of great significance for downstream problems with sentence-pair input. Despite the effectiveness of NSP, we suggest that NSP still lacks the essential signal to distinguish between entailment and shallow correlation. To remedy this, we propose to augment the NSP task to a 3-class categorization task, which includes a category for previous sentence prediction (PSP). The involvement of PSP encourages the model to focus on the informative semantics to determine the sentence order, thereby improves the ability of semantic understanding. This simple modification yields remarkable improvement against vanilla BERT. To further incorporate the document-level information, the scope of NSP and PSP is expanded into a broader range, i.e., NSP and PSP also include close but nonsuccessive sentences, the noise of which is mitigated by the label-smoothing technique. Both qualitative and quantitative experimental results demonstrate the effectiveness of the proposed method. Our method consistently improves the performance on the NLI and MRC benchmarks, including the challenging HANS dataset \cite{hans}, suggesting that the document-level task is still promising for the pre-training.

cs.CL↗

arXiv2021

PairRE: Knowledge Graph Embeddings via Paired Relation Vectors

Distance based knowledge graph embedding methods show promising results on link prediction task, on which two topics have been widely studied: one is the ability to handle complex relations, such as N-to-1, 1-to-N and N-to-N, the other is to encode various relation patterns, such as symmetry/antisymmetry. However, the existing methods fail to solve these two problems at the same time, which leads to unsatisfactory results. To mitigate this problem, we propose PairRE, a model with paired vectors for each relation representation. The paired vectors enable an adaptive adjustment of the margin in loss function to fit for complex relations. Besides, PairRE is capable of encoding three important relation patterns, symmetry/antisymmetry, inverse and composition. Given simple constraints on relation representations, PairRE can encode subrelation further. Experiments on link prediction benchmarks demonstrate the proposed key capabilities of PairRE. Moreover, We set a new state-of-the-art on two knowledge graph datasets of the challenging Open Graph Benchmark.

cs.CL↗

arXiv2021

An electro-optically tunable microring laser monolithically integrated on lithium niobate on insulator

We demonstrate monolithic integration of an electro-optically (EO) tunable microring laser on lithium niobate on insulator (LNOI) platform. The device is fabricated by photolithography assisted chemo-mechanical etching (PLACE), and the pump laser is evanescently coupled into the erbium (Er3+) doped LN microring laser using an undoped LN waveguide mounted above the microring. The quality factor of the LN microring resonator is measured as high as 1.54x10^5 at the wavelength of 1542 nm. Lasing action can be observed at a pump power threshold below 3.5 mW using a 980 nm continuous-wave pump laser. Finally, tuning of the laser wavelength is achieved by varying the electric voltage on the microelectrodes fabricated in the vicinity of microring waveguide, showing an EO coefficient of 0.33 pm/V.

physics.optics↗

arXiv2021

CASS-NAT: CTC Alignment-based Single Step Non-autoregressive Transformer for Speech Recognition

We propose a CTC alignment-based single step non-autoregressive transformer (CASS-NAT) for speech recognition. Specifically, the CTC alignment contains the information of (a) the number of tokens for decoder input, and (b) the time span of acoustics for each token. The information are used to extract acoustic representation for each token in parallel, referred to as token-level acoustic embedding which substitutes the word embedding in autoregressive transformer (AT) to achieve parallel generation in decoder. During inference, an error-based alignment sampling method is proposed to be applied to the CTC output space, reducing the WER and retaining the parallelism as well. Experimental results show that the proposed method achieves WERs of 3.8%/9.1% on Librispeech test clean/other dataset without an external LM, and a CER of 5.8% on Aishell1 Mandarin corpus, respectively1. Compared to the AT baseline, the CASS-NAT has a performance reduction on WER, but is 51.2x faster in terms of RTF. When decoding with an oracle CTC alignment, the lower bound of WER without LM reaches 2.3% on the test-clean set, indicating the potential of the proposed method.

eess.AS↗

arXiv2021

On-chip integrated waveguide amplifiers on Erbium-doped thin film lithium niobate on insulator

We demonstrate on-chip light amplification with integrated optical waveguide fabricated on erbium-doped thin film lithium niobate on insulator (TFLNOI) using the photolithography assisted chemo-mechanical etching (PLACE) technique. A maximum internal net gain of 18 dB in the small-signal-gain regime is measured at the peak emission wavelength of 1530 nm for a waveguide length of 3.6 cm, indicating a differential gain per unit length of 5 dB/cm. This work paves the way to the monolithic integration of diverse active and passive photonic components on the TFLNOI platform.

physics.optics↗

arXiv2020

Toward Fast and Accurate Neural Chinese Word Segmentation with Multi-Criteria Learning

The ambiguous annotation criteria lead to divergence of Chinese Word Segmentation (CWS) datasets in various granularities. Multi-criteria Chinese word segmentation aims to capture various annotation criteria among datasets and leverage their common underlying knowledge. In this paper, we propose a domain adaptive segmenter to exploit diverse criteria of various datasets. Our model is based on Bidirectional Encoder Representations from Transformers (BERT), which is responsible for introducing open-domain knowledge. Private and shared projection layers are proposed to capture domain-specific knowledge and common knowledge, respectively. We also optimize computational efficiency via distillation, quantization, and compiler optimization. Experiments show that our segmenter outperforms the previous state of the art (SOTA) models on 10 CWS datasets with superior efficiency.

cs.CL↗

arXiv2020

An on-chip tunable micro-disk laser fabricated on Er3+ doped lithium niobate on insulator (LNOI)

We demonstrate a C-band wavelength-tunable microlaser with an Er3+ doped high quality (~1.02x10^6) lithium niobate microdisk resonator. With a 976 nm continuous-wave pump laser, lasing action can be observed at a pump power threshold as low as ~250 μW at room temperature. Furthermore, the microdisk laser wavelength can be tuned by varying the pump laser power, showing a tuning efficiency of ~-17.03 pm/mW at low pump power blow 13 mW, and 10.58 pm/mW at high pump power above 13 mW.

physics.optics↗

arXiv2020

Riemannian Proximal Policy Optimization

In this paper, We propose a general Riemannian proximal optimization algorithm with guaranteed convergence to solve Markov decision process (MDP) problems. To model policy functions in MDP, we employ Gaussian mixture model (GMM) and formulate it as a nonconvex optimization problem in the Riemannian space of positive semidefinite matrices. For two given policy functions, we also provide its lower bound on policy improvement by using bounds derived from the Wasserstein distance of GMMs. Preliminary experiments show the efficacy of our proposed Riemannian proximal policy optimization algorithm.

cs.LG↗

arXiv2020

SpellGCN: Incorporating Phonological and Visual Similarities into Language Models for Chinese Spelling Check

Chinese Spelling Check (CSC) is a task to detect and correct spelling errors in Chinese natural language. Existing methods have made attempts to incorporate the similarity knowledge between Chinese characters. However, they take the similarity knowledge as either an external input resource or just heuristic rules. This paper proposes to incorporate phonological and visual similarity knowledge into language models for CSC via a specialized graph convolutional network (SpellGCN). The model builds a graph over the characters, and SpellGCN is learned to map this graph into a set of inter-dependent character classifiers. These classifiers are applied to the representations extracted by another network, such as BERT, enabling the whole network to be end-to-end trainable. Experiments (The dataset and all code for this paper are available at https://github.com/ACL2020SpellGCN/SpellGCN) are conducted on three human-annotated datasets. Our method achieves superior performance against previous models by a large margin.

cs.CL↗

arXiv2020

High-index-contrast single-mode optical waveguides fabricated on lithium niobate by photolithography assisted chemo-mechanical etching (PLACE)

We report fabrication of low loss single mode waveguides on lithium niobate on insulator (LNOI) cladded by a layer of SiO2. Our technique, termed photolithography assisted chemo-mechanical etching (PLACE), relies on patterning of a chromium film into the mask shape by femtosecond laser micromachining and subsequent chemo-mechanical etching of the lithium niobate thin film. The high-index-contrast single mode waveguide is measured to have a propagation loss of 0.13 dB/cm. Furthermore, waveguide tapers are fabricated for boosting the coupling efficiency.

physics.app-ph↗

arXiv2020

Incubation Induced Light Concentration Beyond the Diffraction Limit for High-Resolution Glass Printing

In the past two decades, tremendous efforts have been exerted to understand and control the delivery of ultrashort laser pulses into various types of transparent materials ranging from glass and crystal to polymer and even bio-materials. This approach opens up the route toward determinative and highly localized modification within the transparent materials, enabling three-dimensional (3D) micromachining of the materials into sophisticated structures and devices with the extreme geometrical flexibility. Owing to the linear diffraction and nonlinear self-focusing effects, the focal volume typically exhibits an asymmetric profile stretching along the longitudinal direction. This effect becomes more severe when focusing deeply into the transparent substrates for printing objects of large heights. In this work a new laser-material interaction regime is identified with the exceptional incubation effect originating from self-regulated multiple-pulse interactions with accumulated material changes. Our finding reveals a focal-volume-invariant modification deeply inside the fused silica glass, in striking contrary to the traditional believes that the geometrical shape of the laser induced modification follows the intensity distribution of the inscription laser. A macro-scale geometrically complex glass sculpture is successfully manufactured with the incubation assisted ultrashort laser inscription at uniform micrometer resolutions in all three dimensions.

physics.optics↗

arXiv2019

Extreme nonlinear Raman interaction of an ultrashort nitrogen ion laser with an impulsively excited molecular wavepacket

We report generation of cascaded rotational Raman scattering up to 58th orders in coherently excited CO_2 molecules. The high-order Raman scattering, which produces a quasiperiodic frequency comb with more than 600 sidebands, is obtained using an intense femtosecond laser to impulsively excite rotational coherence and the femtosecond-laser-induced N_2^+ lasing to generate cascaded Raman signals. The novel configuration allows this experiment to be performed with a single femtosecond laser beam at free-space standoff locations. It is revealed that the efficient spectral extension of Raman signals is attributed to the specific spectra-temporal structures of N_2^+ lasing, the ideal spatial overlap of femtosecond laser and N2+ lasing, and the guiding effect of molecular alignment. The Raman spectrum extending above 2000 cm^-1 naturally corresponds to a femtosecond pulse train due to the periodic revivals of molecular rotational wavepackets.

physics.optics↗

arXiv2019

Freeform microfluidic networks encapsulated in laser printed three-dimensional macro-scale glass objects

Large-scale microfluidic microsystems with complex three-dimensional (3D) configurations are highly in demand by both fundamental research and industrial application, holding the potentials for fostering a wide range of innovative applications such as lab-on-a-chip and organ-on-a-chip as well as continuous-flow manufacturing of fine chemicals. However, freeform fabrication of such systems remains challenging for most of the current fabrication techniques in terms of fabrication resolution, flexibility, and achievable footprint size. Here, we report ultrashort pulse laser microfabrication of freeform microfluidic circuits with high aspect ratios and tunable diameters embedded in 3D printed glass objects. We achieve uniform microfluidic channel diameter by carefully distributing a string of extra access ports along the microfluidic channels for avoiding the over-etching in the thin microfluidic channels. After the chemical etching is completed, the extra access ports are sealed using carbon dioxide laser induced localized glass melting. We demonstrate a model hand of fused silica with a size of ~3 cm * 2.7 cm * 1.1 cm in which the whole blood vessel system is encapsulated.

physics.app-ph↗

arXiv2019

A compact and efficient three-dimensional microfluidic mixer

Microfluidic mixing is a fundamental functionality in most lab on a chip (LOC) systems,whereas realization of efficient mixing is challenging in microfluidic channels due to the small Reynolds numbers. Here, we design and fabricate a compact three-dimensional (3D) micromixer to enable efficient mixing at various flow rates. The performance of the fabricated micromixer was examined using blue and red inks. The extreme flexibility in fabricating microfluidic structures of arbitrary 3D geometries using femtosecond laser micromachining allows us to tackle the major disadvantageous effects for optimizing the mixing efficiency.

physics.app-ph↗

arXiv2019

BERT-Based Multi-Head Selection for Joint Entity-Relation Extraction

In this paper, we report our method for the Information Extraction task in 2019 Language and Intelligence Challenge. We incorporate BERT into the multi-head selection framework for joint entity-relation extraction. This model extends existing approaches from three perspectives. First, BERT is adopted as a feature extraction layer at the bottom of the multi-head selection framework. We further optimize BERT by introducing a semantic-enhanced task during BERT pre-training. Second, we introduce a large-scale Baidu Baike corpus for entity recognition pre-training, which is of weekly supervised learning since there is no actual named entity label. Third, soft label embedding is proposed to effectively transmit information between entity recognition and relation extraction. Combining these three contributions, we enhance the information extracting ability of the multi-head selection model and achieve F1-score 0.876 on testset-1 with a single model. By ensembling four variants of our model, we finally achieve F1 score 0.892 (1st place) on testset-1 and F1 score 0.8924 (2nd place) on testset-2.

cs.CL↗

arXiv2019

Variational Semi-supervised Aspect-term Sentiment Analysis via Transformer

Aspect-term sentiment analysis (ATSA) is a longstanding challenge in natural language understanding. It requires fine-grained semantical reasoning about a target entity appeared in the text. As manual annotation over the aspects is laborious and time-consuming, the amount of labeled data is limited for supervised learning. This paper proposes a semi-supervised method for the ATSA problem by using the Variational Autoencoder based on Transformer (VAET), which models the latent distribution via variational inference. By disentangling the latent representation into the aspect-specific sentiment and the lexical context, our method induces the underlying sentiment prediction for the unlabeled data, which then benefits the ATSA classifier. Our method is classifier agnostic, i.e., the classifier is an independent module and various advanced supervised models can be integrated. Experimental results are obtained on the SemEval 2014 task 4 and show that our method is effective with four classical classifiers. The proposed method outperforms two general semisupervised methods and achieves state-of-the-art performance.

cs.CL↗

arXiv2019

Efficient Electro-optical Tuning of Optical Frequency Microcomb on a Monolithically Integrated High-Q Lithium Niobate Microdisk

We demonstrate efficient tuning of a monolithically integrated lithium niobate microdisk (LN) optical frequency microcomb. Utilizing the high optical quality (Q) factor (i.e., Q~7.1*10^6) of the microdisk, the microcomb spans over a spectral bandwidth of ~200 nm at a pump power as low as 20.4 mW. Combining the large eletro-optic coefficient of LN and optimum design of the geometry of microelectrodes, we demonstrate electro-optical tuning of the comb with a spectral range of 400 pm and a tuning efficiency of ~38 pm/100V.

physics.optics↗