arXiv subjects

Hangting Chen

Publications and source records attributed to Hangting Chen.

35 records · Page 2Linked to original sources

arXiv2024

Continuous Target Speech Extraction: Enhancing Personalized Diarization and Extraction on Complex Recordings

Target speaker extraction (TSE) aims to extract the target speaker's voice from the input mixture. Previous studies have concentrated on high-overlapping scenarios. However, real-world applications usually meet more complex scenarios like variable speaker overlapping and target speaker absence. In this paper, we introduces a framework to perform continuous TSE (C-TSE), comprising a target speaker voice activation detection (TSVAD) and a TSE model. This framework significantly improves TSE performance on similar speakers and enhances personalization, which is lacking in traditional diarization methods. In detail, unlike conventional TSVAD deployed to refine the diarization results, the proposed Attention-target speaker voice activation detection (A-TSVAD) directly generates timestamps of the target speaker. We also explore some different integration methods of A-TSVAD and TSE by comparing the cascaded and parallel methods. The framework's effectiveness is assessed using a range of metrics, including diarization and enhancement metrics. Our experiments demonstrate that A-TSVAD outperforms conventional methods in reducing diarization errors. Furthermore, the integration of A-TSVAD and TSE in a sequential cascaded manner further enhances extraction accuracy.

cs.SD↗

arXiv2023

Consistent and Relevant: Rethink the Query Embedding in General Sound Separation

The query-based audio separation usually employs specific queries to extract target sources from a mixture of audio signals. Currently, most query-based separation models need additional networks to obtain query embedding. In this way, separation model is optimized to be adapted to the distribution of query embedding. However, query embedding may exhibit mismatches with separation models due to inconsistent structures and independent information. In this paper, we present CaRE-SEP, a consistent and relevant embedding network for general sound separation to encourage a comprehensive reconsideration of query usage in audio separation. CaRE-SEP alleviates the potential mismatch between queries and separation in two aspects, including sharing network structure and sharing feature information. First, a Swin-Unet model with a shared encoder is conducted to unify query encoding and sound separation into one model, eliminating the network architecture difference and generating consistent distribution of query and separation features. Second, by initializing CaRE-SEP with a pretrained classification network and allowing gradient backpropagation, the query embedding is optimized to be relevant to the separation feature, further alleviating the feature mismatch problem. Experimental results indicate the proposed CaRE-SEP model substantially improves the performance of separation tasks. Moreover, visualizations validate the potential mismatch and how CaRE-SEP solves it.

eess.AS↗

arXiv2023

SECap: Speech Emotion Captioning with Large Language Model

Speech emotions are crucial in human communication and are extensively used in fields like speech synthesis and natural language understanding. Most prior studies, such as speech emotion recognition, have categorized speech emotions into a fixed set of classes. Yet, emotions expressed in human speech are often complex, and categorizing them into predefined groups can be insufficient to adequately represent speech emotions. On the contrary, describing speech emotions directly by means of natural language may be a more effective approach. Regrettably, there are not many studies available that have focused on this direction. Therefore, this paper proposes a speech emotion captioning framework named SECap, aiming at effectively describing speech emotions using natural language. Owing to the impressive capabilities of large language models in language comprehension and text generation, SECap employs LLaMA as the text decoder to allow the production of coherent speech emotion captions. In addition, SECap leverages HuBERT as the audio encoder to extract general speech features and Q-Former as the Bridge-Net to provide LLaMA with emotion-related speech features. To accomplish this, Q-Former utilizes mutual information learning to disentangle emotion-related speech features and speech contents, while implementing contrastive learning to extract more emotion-related speech features. The results of objective and subjective evaluations demonstrate that: 1) the SECap framework outperforms the HTSAT-BART baseline in all objective evaluations; 2) SECap can generate high-quality speech emotion captions that attain performance on par with human annotators in subjective mean opinion score tests.

cs.SD↗

arXiv2023

Ultra Dual-Path Compression For Joint Echo Cancellation And Noise Suppression

Echo cancellation and noise reduction are essential for full-duplex communication, yet most existing neural networks have high computational costs and are inflexible in tuning model complexity. In this paper, we introduce time-frequency dual-path compression to achieve a wide range of compression ratios on computational cost. Specifically, for frequency compression, trainable filters are used to replace manually designed filters for dimension reduction. For time compression, only using frame skipped prediction causes large performance degradation, which can be alleviated by a post-processing network with full sequence modeling. We have found that under fixed compression ratios, dual-path compression combining both the time and frequency methods will give further performance improvement, covering compression ratios from 4x to 32x with little model size change. Moreover, the proposed models show competitive performance compared with fast FullSubNet and DeepFilterNet.

eess.AS↗

arXiv2023

AutoPrep: An Automatic Preprocessing Framework for In-the-Wild Speech Data

Recently, the utilization of extensive open-sourced text data has significantly advanced the performance of text-based large language models (LLMs). However, the use of in-the-wild large-scale speech data in the speech technology community remains constrained. One reason for this limitation is that a considerable amount of the publicly available speech data is compromised by background noise, speech overlapping, lack of speech segmentation information, missing speaker labels, and incomplete transcriptions, which can largely hinder their usefulness. On the other hand, human annotation of speech data is both time-consuming and costly. To address this issue, we introduce an automatic in-the-wild speech data preprocessing framework (AutoPrep) in this paper, which is designed to enhance speech quality, generate speaker labels, and produce transcriptions automatically. The proposed AutoPrep framework comprises six components: speech enhancement, speech segmentation, speaker clustering, target speech extraction, quality filtering and automatic speech recognition. Experiments conducted on the open-sourced WenetSpeech and our self-collected AutoPrepWild corpora demonstrate that the proposed AutoPrep framework can generate preprocessed data with similar DNSMOS and PDNSMOS scores compared to several open-sourced TTS datasets. The corresponding TTS system can achieve up to 0.68 in-domain speaker similarity.

eess.AS↗

arXiv2023

Complexity Scaling for Speech Denoising

Computational complexity is critical when deploying deep learning-based speech denoising models for on-device applications. Most prior research focused on optimizing model architectures to meet specific computational cost constraints, often creating distinct neural network architectures for different complexity limitations. This study conducts complexity scaling for speech denoising tasks, aiming to consolidate models with various complexities into a unified architecture. We present a Multi-Path Transform-based (MPT) architecture to handle both low- and high-complexity scenarios. A series of MPT networks present high performance covering a wide range of computational complexities on the DNS challenge dataset. Moreover, inspired by the scaling experiments in natural language processing, we explore the empirical relationship between model performance and computational cost on the denoising task. As the complexity number of multiply-accumulate operations (MACs) is scaled from 50M/s to 15G/s on MPT networks, we observe a linear increase in the values of PESQ-WB and SI-SNR, proportional to the logarithm of MACs, which might contribute to the understanding and application of complexity scaling in speech denoising tasks.

eess.AS↗

arXiv2023

Bayes Risk Transducer: Transducer with Controllable Alignment Prediction

Automatic speech recognition (ASR) based on transducers is widely used. In training, a transducer maximizes the summed posteriors of all paths. The path with the highest posterior is commonly defined as the predicted alignment between the speech and the transcription. While the vanilla transducer does not have a prior preference for any of the valid paths, this work intends to enforce the preferred paths and achieve controllable alignment prediction. Specifically, this work proposes Bayes Risk Transducer (BRT), which uses a Bayes risk function to set lower risk values to the preferred paths so that the predicted alignment is more likely to satisfy specific desired properties. We further demonstrate that these predicted alignments with intentionally designed properties can provide practical advantages over the vanilla transducer. Experimentally, the proposed BRT saves inference cost by up to 46% for non-streaming ASR and reduces overall system latency by 41% for streaming ASR.

cs.CL↗

arXiv2023

High Fidelity Speech Enhancement with Band-split RNN

Despite the rapid progress in speech enhancement (SE) research, enhancing the quality of desired speech in environments with strong noise and interfering speakers remains challenging. In this paper, we extend the application of the recently proposed band-split RNN (BSRNN) model to full-band SE and personalized SE (PSE) tasks. To mitigate the effects of unstable high-frequency components in full-band speech, we perform bi-directional and uni-directional band-level modeling to low-frequency and high-frequency subbands, respectively. For PSE task, we incorporate a speaker enrollment module into BSRNN to utilize target speaker information. Moreover, we utilize a MetricGAN discriminator (MGD) and a multi-resolution spectrogram discriminator (MRSD) to improve perceptual quality metrics. Experimental results show that our system outperforms various top-ranking SE systems, achieves state-of-the-art (SOTA) results on the DNS-2020 test set and ranks among the top 3 in the DNS-2023 challenge.

eess.AS↗

arXiv2022

The HCCL System for the NIST SRE21

This paper describes the systems developed by the HCCL team for the NIST 2021 speaker recognition evaluation (NIST SRE21).We first explore various state-of-the-art speaker embedding extractors combined with a novel circle loss to obtain discriminative deep speaker embeddings. Considering that cross-channel and cross-linguistic speaker recognition are the key challenges of SRE21, we introduce several techniques to reduce the cross-domain mismatch. Specifically, Codec and speech enhancement are directly applied to the raw speech to eliminate the codecs and the environment noise mismatch. We denote the methods that work directly on speech to eliminate the relatively explicit mismatches collectively as data adaptation methods. Experiments show that data adaption methods achieve 15\% improvements over our baseline. Furthermore, some popular back-ends domain adaptation algorithms are deployed on speaker embeddings to alleviate speaker performance degradation caused by the implicit mismatch. Score calibration is a major failure for us in SRE21. The reason is that score calibration with too many parameters easily lead to overfitting problems.

cs.SD↗

arXiv2022

Beam-Guided TasNet: An Iterative Speech Separation Framework with Multi-Channel Output

Time-domain audio separation network (TasNet) has achieved remarkable performance in blind source separation (BSS). Classic multi-channel speech processing framework employs signal estimation and beamforming. For example, Beam-TasNet links multi-channel convolutional TasNet (MC-Conv-TasNet) with minimum variance distortionless response (MVDR) beamforming, which leverages the strong modeling ability of data-driven network and boosts the performance of beamforming with an accurate estimation of speech statistics. Such integration can be viewed as a directed acyclic graph by accepting multi-channel input and generating multi-source output. In this paper, we design a "multi-channel input, multi-channel multi-source output" (MIMMO) speech separation system entitled "Beam-Guided TasNet", where MC-Conv-TasNet and MVDR can interact and promote each other more compactly under a directed cyclic flow. Specifically, the first stage uses Beam-TasNet to generate estimated single-speaker signals, which favors the separation in the second stage. The proposed framework facilitates iterative signal refinement with the guide of beamforming and seeks to reach the upper bound of the MVDR-based methods. Experimental results on the spatialized WSJ0-2MIX demonstrate that the Beam-Guided TasNet has achieved an SDR of 21.5 dB, exceeding the baseline Beam-TasNet by 4.1 dB under the same model size and narrowing the gap with the oracle signal-based MVDR to 2 dB.

eess.AS↗

arXiv2022

DPT-FSNet: Dual-path Transformer Based Full-band and Sub-band Fusion Network for Speech Enhancement

Sub-band models have achieved promising results due to their ability to model local patterns in the spectrogram. Some studies further improve the performance by fusing sub-band and full-band information. However, the structure for the full-band and sub-band fusion model was not fully explored. This paper proposes a dual-path transformer-based full-band and sub-band fusion network (DPT-FSNet) for speech enhancement in the frequency domain. The intra and inter parts of the dual-path transformer model sub-band and full-band information, respectively. The features utilized by our proposed method are more interpretable than those utilized by the time-domain dual-path transformer. We conducted experiments on the Voice Bank + DEMAND and Interspeech 2020 Deep Noise Suppression (DNS) datasets to evaluate the proposed method. Experimental results show that the proposed method outperforms the current state-of-the-art.

cs.SD↗

arXiv2021

Exploring the time-domain deep attractor network with two-stream architectures in a reverberant environment

Deep attractor networks (DANs) perform speech separation with discriminative embeddings and speaker attractors. Compared with methods based on the permutation invariant training (PIT), DANs define a deep embedding space and deliver a more elaborate representation on each time-frequency (T-F) bin. However, it has been observed that the DANs achieve limited improvement on the signal quality if directly deployed in a reverberant environment. Following the success of time-domain separation networks on the clean mixture speech, we propose a time-domain DAN (TD-DAN) with two-streams of convolutional networks, which efficiently perform both dereverberation and separation tasks under the condition of a variable number of speakers. The speaker encoding stream (SES) of the TD-DAN is trained to model the speaker information in the embedding space. The speech decoding stream (SDS) accepts speaker attractors from the SES and learns to estimate early reflections from the spectro-temporal representations. Meanwhile, additional clustering losses are used to bridge the gap between the oracle and the estimated attractors. Experiments were conducted on the Spatialized Multi-Speaker Wall Street Journal (SMS-WSJ) dataset. The early reflection was compared with the anechoic and reverberant signals and then was chosen as the learning targets. The experimental results demonstrated that the TD-DAN achieved scale-invariant source-to-distortion ratio (SI-SDR) gains of 9.79/7.47 dB on the reverberant 2/3-speaker evaluation set, exceeding the baseline DAN and convolutional time-domain audio separation network (Conv-TasNet) by 1.92/0.68 dB and 0.91/0.47 dB, respectively.

eess.AS↗

arXiv2021

Power pooling: An adaptive pooling function for weakly labelled sound event detection

Access to large corpora with strongly labelled sound events is expensive and difficult in engineering applications. Much research turns to address the problem of how to detect both the types and the timestamps of sound events with weak labels that only specify the types. This task can be treated as a multiple instance learning (MIL) problem, and the key to it is the design of a pooling function. In this paper, we propose an adaptive power pooling function which can automatically adapt to various sound sources. On two public datasets, the proposed power pooling function outperforms the state-of-the-art linear softmax pooling on both coarsegrained and fine-grained metrics. Notably, it improves the event-based F1 score (which evaluates the detection of event onsets and offsets) by 11.4% and 10.2% relative on the two datasets. While this paper focuses on sound event detection applications, the proposed method can be applied to MIL tasks in other domains.

cs.SD↗

arXiv2020

ACGAN-based Data Augmentation Integrated with Long-term Scalogram for Acoustic Scene Classification

In acoustic scene classification (ASC), acoustic features play a crucial role in the extraction of scene information, which can be stored over different time scales. Moreover, the limited size of the dataset may lead to a biased model with a poor performance for records from unseen cities and confusing scene classes. In order to overcome this, we propose a long-term wavelet feature that requires a lower storage capacity and can be classified faster and more accurately compared with classic Mel filter bank coefficients (FBank). This feature can be extracted with predefined wavelet scales similar to the FBank. Furthermore, a novel data augmentation scheme based on generative adversarial neural networks with auxiliary classifiers (ACGANs) is adopted to improve the generalization of the ASC systems. The scheme, which contains ACGANs and a sample filter, extends the database iteratively by splitting the dataset, training the ACGANs and subsequently filtering samples. Experiments were conducted on datasets from the Detection and Classification of Acoustic Scenes and Events (DCASE) challenges. The results on the DCASE19 dataset demonstrate the improved performance of the proposed techniques compared with the classic FBank classifier. Moreover, the proposed fusion system achieved first place in the DCASE19 competition and surpassed the top accuracies on the DCASE17 dataset.

eess.AS↗

arXiv2020

Power Pooling Operators and Confidence Learning for Semi-Supervised Sound Event Detection

In recent years, the involvement of synthetic strongly labeled data,weakly labeled data and unlabeled data has drawn much research attentionin semi-supervised sound event detection (SSED). Self-training models carry out predictions without strong annotations and then take predictions with high probabilities as pseudo-labels for retraining. Such models have shown its effectiveness in SSED. However, probabilities are poorly calibrated confidence estimates, and samples with low probabilities are ignored. Hence, we introduce a method of learning confidence deliberately and retaining all data distinctly by applying confidence as weights. Additionally, linear pooling has been considered as a state-of-the-art aggregation function for SSED with weak labeling. In this paper, we propose a power pooling function whose coefficient can be trained automatically to achieve nonlinearity. A confidencebased semi-supervised sound event detection (C-SSED) framework is designed to combine confidence and power pooling. The experimental results demonstrate that confidence is proportional to the accuracy of the predictions. The power pooling function outperforms linear pooling at both error rate and F1 results. In addition, the C-SSED framework achieves a relative error rate reduction of 34% in contrast to the baseline model.

cs.SD↗

arXiv2019

Multi-Talker MVDR Beamforming Based on Extended Complex Gaussian Mixture Model

In this letter, we present a novel multi-talker minimum variance distortionless response (MVDR) beamforming as the front-end of an automatic speech recognition (ASR) system in a dinner party scenario. The CHiME-5 dataset is selected to evaluate our proposal for overlapping multi-talker scenario with severe noise. A detailed study on beamforming is conducted based on the proposed extended complex Gaussian mixture model (CGMM) integrated with various speech separation and speech enhancement masks. Three main changes are made to adopt the original CGMM-based MVDR for the multi-talker scenario. First, the number of Gaussian distributions is extended to 3 with an additional inference speaker model. Second, the mixture coefficients are introduced as a supervisor to generate more elaborate masks and avoid the permutation problems. Moreover, we reorganize the MVDR and mask-based speech separation to achieve both noise reduction and target speaker extraction. With the official baseline ASR back-end, our front-end algorithm gained an absolute WER reduction of 13.87% compared with the baseline front-end.

eess.AS↗

arXiv2019

Integrating the Data Augmentation Scheme with Various Classifiers for Acoustic Scene Modeling

This technical report describes the IOA team's submission for TASK1A of DCASE2019 challenge. Our acoustic scene classification (ASC) system adopts a data augmentation scheme employing generative adversary networks. Two major classifiers, 1D deep convolutional neural network integrated with scalogram features and 2D fully convolutional neural network integrated with Mel filter bank features, are deployed in the scheme. Other approaches, such as adversary city adaptation, temporal module based on discrete cosine transform and hybrid architectures, have been developed for further fusion. The results of our experiments indicates that the final fusion systems A-D could achieve an accuracy higher than 85% on the officially provided fold 1 evaluation dataset.

eess.AS↗