arXiv subjects

He Huang

Publications and source records attributed to He Huang.

At least 73 records · Page 4Linked to original sources

arXiv2024

DeSTA: Enhancing Speech Language Models through Descriptive Speech-Text Alignment

Recent speech language models (SLMs) typically incorporate pre-trained speech models to extend the capabilities from large language models (LLMs). In this paper, we propose a Descriptive Speech-Text Alignment approach that leverages speech captioning to bridge the gap between speech and text modalities, enabling SLMs to interpret and generate comprehensive natural language descriptions, thereby facilitating the capability to understand both linguistic and non-linguistic features in speech. Enhanced with the proposed approach, our model demonstrates superior performance on the Dynamic-SUPERB benchmark, particularly in generalizing to unseen tasks. Moreover, we discover that the aligned model exhibits a zero-shot instruction-following capability without explicit speech instruction tuning. These findings highlight the potential to reshape instruction-following SLMs by incorporating rich, descriptive speech captions.

eess.AS↗

arXiv2023

Infrared anomalies in ultrathin Ti3C2Tx MXene films

Visible transparent but infrared reflective materials are ideal candidates for both transparent conductive films and low-emissivity glass, which are highly desired in a broad variety of areas such as touchscreens and displays, photovoltaics, smart windows, and antistatic coatings. Ultrathin Ti3C2Tx MXene films are emerging as promising low-emissivity transparent candidates. However, the fundamental IR properties of Ti3C2Tx has not been revealed experimentally due to daunting challenges in the preparation of continuous, large-area, and ultrathin films of optical quality on flat substrates. Herein, we proposed a tape-free transfer method that can help prepare centimeter-size and ultrathin (down to 8 nm) Ti3C2Tx films on diverse optical substrates. Benefitting from this method, the refractive index and permittivity for Ti3C2Tx were successfully measured. Ti3C2Tx films exhibit large in-plane permittivity in the IR region, yielding maximum IR reflectance of 88% for bulk films. Interestingly, three anomalies were found in ultrathin Ti3C2Tx films: strong dispersion in the permittivity, interlayer space-dependent optical properties, and abnormally high IR absorption for a 15-nm-thick film. These anomalies are important guidelines in the design of Ti3C2Tx-based low-emissivity transparent films and other related devices, and may inspire other intriguing applications such as ultrathin IR absorption coatings and tunable IR optical devices.

cond-mat.mtrl-sci↗

arXiv2023

Correlated two-photon scattering in a one-dimensional waveguide coupled to two- or three-level giant atoms

We study the two-photon scattering processes in a one-dimensional waveguide coupled to a two- or three-level giant atom, respectively. The accumulated phase shift between the two coupling points can be utilized to alter the scattering processes. We obtain the exact interacting two-photon scattering wavefunction of these two systems following the Lippmann-Schwinger formalism, from which the analytical expressions of incoherent power spectra and second-order correlations are also derived. The incoherent spectrum, defined by the correlation of the bound state, serves as a useful indication of photon-photon correlation. The second-order correlation function gives a direct measure of photon-photon correlation. For photons scattered by the two-level giant atom, the accumulated phase shift can be used to improve photon-photon correlation,and adjust the evolution of the second-order correlation. In the system of the three-level giant atom, the photon-photon correlation can be substantially increased. Moreover, the photon-photon interactions and correlation distance of scattered photons can be further enhanced by tuning the accumulated phase shift.

quant-ph↗

arXiv2023

Property-Aware Multi-Speaker Data Simulation: A Probabilistic Modelling Technique for Synthetic Data Generation

We introduce a sophisticated multi-speaker speech data simulator, specifically engineered to generate multi-speaker speech recordings. A notable feature of this simulator is its capacity to modulate the distribution of silence and overlap via the adjustment of statistical parameters. This capability offers a tailored training environment for developing neural models suited for speaker diarization and voice activity detection. The acquisition of substantial datasets for speaker diarization often presents a significant challenge, particularly in multi-speaker scenarios. Furthermore, the precise time stamp annotation of speech data is a critical factor for training both speaker diarization and voice activity detection. Our proposed multi-speaker simulator tackles these problems by generating large-scale audio mixtures that maintain statistical properties closely aligned with the input parameters. We demonstrate that the proposed multi-speaker simulator generates audio mixtures with statistical properties that closely align with the input parameters derived from real-world statistics. Additionally, we present the effectiveness of speaker diarization and voice activity detection models, which have been trained exclusively on the generated simulated datasets.

eess.AS↗

arXiv2023

The CHiME-7 Challenge: System Description and Performance of NeMo Team's DASR System

We present the NVIDIA NeMo team's multi-channel speech recognition system for the 7th CHiME Challenge Distant Automatic Speech Recognition (DASR) Task, focusing on the development of a multi-channel, multi-speaker speech recognition system tailored to transcribe speech from distributed microphones and microphone arrays. The system predominantly comprises of the following integral modules: the Speaker Diarization Module, Multi-channel Audio Front-End Processing Module, and the ASR Module. These components collectively establish a cascading system, meticulously processing multi-channel and multi-speaker audio input. Moreover, this paper highlights the comprehensive optimization process that significantly enhanced our system's performance. Our team's submission is largely based on NeMo toolkits and will be publicly available.

eess.AS↗

arXiv2023

SALM: Speech-augmented Language Model with In-context Learning for Speech Recognition and Translation

We present a novel Speech Augmented Language Model (SALM) with {\em multitask} and {\em in-context} learning capabilities. SALM comprises a frozen text LLM, a audio encoder, a modality adapter module, and LoRA layers to accommodate speech input and associated task instructions. The unified SALM not only achieves performance on par with task-specific Conformer baselines for Automatic Speech Recognition (ASR) and Speech Translation (AST), but also exhibits zero-shot in-context learning capabilities, demonstrated through keyword-boosting task for ASR and AST. Moreover, {\em speech supervised in-context training} is proposed to bridge the gap between LLM training and downstream speech tasks, which further boosts the in-context learning ability of speech-to-text models. Proposed model is open-sourced via NeMo toolkit.

cs.CL↗

arXiv2023

Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition

Conformer-based models have become the dominant end-to-end architecture for speech processing tasks. With the objective of enhancing the conformer architecture for efficient training and inference, we carefully redesigned Conformer with a novel downsampling schema. The proposed model, named Fast Conformer(FC), is 2.8x faster than the original Conformer, supports scaling to Billion parameters without any changes to the core architecture and also achieves state-of-the-art accuracy on Automatic Speech Recognition benchmarks. To enable transcription of long-form speech up to 11 hours, we replaced global attention with limited context attention post-training, while also improving accuracy through fine-tuning with the addition of a global token. Fast Conformer, when combined with a Transformer decoder also outperforms the original Conformer in accuracy and in speed for Speech Translation and Spoken Language Understanding.

eess.AS↗

arXiv2023

MM-Fi: Multi-Modal Non-Intrusive 4D Human Dataset for Versatile Wireless Sensing

4D human perception plays an essential role in a myriad of applications, such as home automation and metaverse avatar simulation. However, existing solutions which mainly rely on cameras and wearable devices are either privacy intrusive or inconvenient to use. To address these issues, wireless sensing has emerged as a promising alternative, leveraging LiDAR, mmWave radar, and WiFi signals for device-free human sensing. In this paper, we propose MM-Fi, the first multi-modal non-intrusive 4D human dataset with 27 daily or rehabilitation action categories, to bridge the gap between wireless sensing and high-level human perception tasks. MM-Fi consists of over 320k synchronized frames of five modalities from 40 human subjects. Various annotations are provided to support potential sensing tasks, e.g., human pose estimation and action recognition. Extensive experiments have been conducted to compare the sensing capacity of each or several modalities in terms of multiple tasks. We envision that MM-Fi can contribute to wireless sensing research with respect to action recognition, human pose estimation, multi-modal learning, cross-modal supervision, and interdisciplinary healthcare research.

eess.SP↗

arXiv2023

Identification of the melting line in the two-dimensional complex plasmas using an unsupervised machine learning method

Machine learning methods have been widely used in the investigations of the complex plasmas. In this paper, we demonstrate that the unsupervised convolutional neural network can be applied to obtain the melting line in the two-dimensional complex plasmas based on the Langevin dynamics simulation results. The training samples do not need to be labeled. The resulting melting line coincides with those obtained by the analysis of hexatic order parameter and supervised machine learning method.

physics.plasm-ph↗

arXiv2023

Leveraging Pretrained ASR Encoders for Effective and Efficient End-to-End Speech Intent Classification and Slot Filling

We study speech intent classification and slot filling (SICSF) by proposing to use an encoder pretrained on speech recognition (ASR) to initialize an end-to-end (E2E) Conformer-Transformer model, which achieves the new state-of-the-art results on the SLURP dataset, with 90.14% intent accuracy and 82.27% SLURP-F1. We compare our model with encoders pretrained on self-supervised learning (SSL), and show that ASR pretraining is much more effective than SSL for SICSF. To explore parameter efficiency, we freeze the encoder and add Adapter modules, and show that parameter efficiency is only achievable with an ASR-pretrained encoder, while the SSL encoder needs full finetuning to achieve comparable results. In addition, we provide an in-depth comparison on end-to-end models versus cascading models (ASR+NLU), and show that E2E models are better than cascaded models unless an oracle ASR model is provided. Last but not least, our model is the first E2E model that achieves the same performance as cascading models with oracle ASR. Code, checkpoints and configs are available.

cs.CL↗

arXiv2023

CMB power spectrum for emergent scenario and slow expansion in scalar-tensor theory of gravity

We analyze the stability of the Einstein static universe in scalar-tensor theory of gravity, and find it can be stable against both scalar and tensor perturbations under certain conditions. By assuming the emergent scenario originating from an Einstein static state, followed by an instantaneous transition to an inflationary phase, we study and obtain the analytical approximations of the primordial power spectrum for the emergent scenario. Then, we plot the primordial power spectrum and CMB TT-spectrum of the emergent scenario and the slow expansion scenario. These figures show that both of these spectra for the slow expansion scenario are the same as that for $Λ$CDM, and the spectra of the emergent scenario are lower than that for $Λ$CDM at large scales.

gr-qc↗

arXiv2023

Efficient Sequence Transduction by Jointly Predicting Tokens and Durations

This paper introduces a novel Token-and-Duration Transducer (TDT) architecture for sequence-to-sequence tasks. TDT extends conventional RNN-Transducer architectures by jointly predicting both a token and its duration, i.e. the number of input frames covered by the emitted token. This is achieved by using a joint network with two outputs which are independently normalized to generate distributions over tokens and durations. During inference, TDT models can skip input frames guided by the predicted duration output, which makes them significantly faster than conventional Transducers which process the encoder output frame by frame. TDT models achieve both better accuracy and significantly faster inference than conventional Transducers on different sequence transduction tasks. TDT models for Speech Recognition achieve better accuracy and up to 2.82X faster inference than conventional Transducers. TDT models for Speech Translation achieve an absolute gain of over 1 BLEU on the MUST-C test compared with conventional Transducers, and its inference is 2.27X faster. In Speech Intent Classification and Slot Filling tasks, TDT models improve the intent accuracy by up to over 1% (absolute) over conventional Transducers, while running up to 1.28X faster. Our implementation of the TDT model will be open-sourced with the NeMo (https://github.com/NVIDIA/NeMo) toolkit.

eess.AS↗

arXiv2023

CMB power spectrum in the emergent universe with k-essence

The emergent universe provides a possible method to avoid the big bang singularity by considering that the universe stems from an stable Einstein static universe rather than the singularity. Since the Einstein static universe exists before inflation, it may leave some relics in the CMB power spectrum. In this paper, we analyze the stability condition for the Einstein static universe in general relativity with k-essence against both the scalar and tensor perturbations. And we find the emergent universe can be successfully realized by constructing a scalar potential and an equation of state parameter. Solving the curved Mukhanov-Sasaki equation, we obtain the analytical approximation for the primordial power spectrum, and then depict the TT-spectrum of the emergent universe. The results show that both the primordial power spectrum and CMB TT-spectrum are suppressed on large scales.

gr-qc↗

arXiv2022

Towards Natural Language-Based Visualization Authoring

A key challenge to visualization authoring is the process of getting familiar with the complex user interfaces of authoring tools. Natural Language Interface (NLI) presents promising benefits due to its learnability and usability. However, supporting NLIs for authoring tools requires expertise in natural language processing, while existing NLIs are mostly designed for visual analytic workflow. In this paper, we propose an authoring-oriented NLI pipeline by introducing a structured representation of users' visualization editing intents, called editing actions, based on a formative study and an extensive survey on visualization construction tools. The editing actions are executable, and thus decouple natural language interpretation and visualization applications as an intermediate layer. We implement a deep learning-based NL interpreter to translate NL utterances into editing actions. The interpreter is reusable and extensible across authoring tools. The authoring tools only need to map the editing actions into tool-specific operations. To illustrate the usages of the NL interpreter, we implement an Excel chart editor and a proof-of-concept authoring tool, VisTalk. We conduct a user study with VisTalk to understand the usage patterns of NL-based authoring systems. Finally, we discuss observations on how users author charts with natural language, as well as implications for future research.

cs.HC↗

arXiv2022

AniVis: Generating Animated Transitions Between Statistical Charts with a Tree Model

Animated transitions help viewers understand changes between related visualizations. To clearly present the underlying relations between statistical charts, animation authors need to have a high level of expertise and a considerable amount of time to describe the relations with reasonable animation stages. We present AniVis, an automated approach for generating animated transitions to demonstrate the changes between two statistical charts. AniVis models each statistical chart into a tree-based structure. Given an input chart pair, the differences of data and visual properties of the chart pair are formalized as tree edit operations. The edit operations can be mapped to atomic transition units. Through this approach, the animated transition between two charts can be expressed as a set of transition units. Then, we conduct a formative study to understand people's preferences for animation sequences. Based on the study, we propose a set of principles and a sequence composition algorithm to compose the transition units into a meaningful animation sequence. Finally, we synthesize these units together to deliver a smooth and intuitive animated transition between charts. To test our approach, we present a prototype system and its generated results to illustrate the usage of our framework. We perform a comparative study to assess the transition sequence derived from the tree model. We further collect qualitative feedback to evaluate the effectiveness and usefulness of our method.

cs.HC↗

arXiv2022

MetaFi: Device-Free Pose Estimation via Commodity WiFi for Metaverse Avatar Simulation

Avatar refers to a representative of a physical user in the virtual world that can engage in different activities and interact with other objects in metaverse. Simulating the avatar requires accurate human pose estimation. Though camera-based solutions yield remarkable performance, they encounter the privacy issue and degraded performance caused by varying illumination, especially in smart home. In this paper, we propose a WiFi-based IoT-enabled human pose estimation scheme for metaverse avatar simulation, namely MetaFi. Specifically, a deep neural network is designed with customized convolutional layers and residual blocks to map the channel state information to human pose landmarks. It is enforced to learn the annotations from the accurate computer vision model, thus achieving cross-modal supervision. WiFi is ubiquitous and robust to illumination, making it a feasible solution for avatar applications in smart home. The experiments are conducted in the real world, and the results show that the MetaFi achieves very high performance with a PCK@50 of 95.23%.

cs.CV↗

arXiv2022

Machine learning in the study of phase transition of two-dimensional complex plasmas

Machine learning is applied to investigate the phase transition of two-dimensional complex plasmas. The Langevin dynamics simulation is employed to prepare particle suspensions in various thermodynamic states. Based on the resulted particle positions in two extreme conditions, bitmap images are synthesized and imported to a convolutional neural network (ConvNet) as training sample. As a result, a phase diagram is obtained. This trained ConvNet model can be directly applied to the sequence of the recorded images using video microscopy in the experiments to study the melting.

physics.plasm-ph↗

arXiv2022

ME-Net: Multi-Encoder Net Framework for Brain Tumor Segmentation

Glioma is the most common and aggressive brain tumor. Magnetic resonance imaging (MRI) plays a vital role to evaluate tumors for the arrangement of tumor surgery and the treatment of subsequent procedures. However, the manual segmentation of the MRI image is strenuous, which limits its clinical application. With the development of deep learning, a large number of automatic segmentation methods have been developed, but most of them stay in 2D images, which leads to subpar performance. Moreover, the serious voxel imbalance between the brain tumor and the background as well as the different sizes and locations of the brain tumor makes the segmentation of 3D images a challenging problem. Aiming at segmenting 3D MRI, we propose a model for brain tumor segmentation with multiple encoders. The structure contains four encoders and one decoder. The four encoders correspond to the four modalities of the MRI image, perform one-to-one feature extraction, and then merge the feature maps of the four modalities into the decoder. This method reduces the difficulty of feature extraction and greatly improves model performance. We also introduced a new loss function named "Categorical Dice", and set different weights for different segmented regions at the same time, which solved the problem of voxel imbalance. We evaluated our approach using the online BraTS 2020 Challenge verification. Our proposed method can achieve promising results in the validation set compared to the state-of-the-art approaches with Dice scores of 0.70249, 0.88267, and 0.73864 for the intact tumor, tumor core, and enhanced tumor, respectively.

eess.IV↗