Xugang Lu

Peng Shen, Xugang Lu, Hisashi Kawai

The acoustic and linguistic features are important cues for the spoken language identification (LID) task. Recent advanced LID systems mainly use acoustic features that lack the usage of explicit linguistic feature encoding. In this paper, we propose a novel transducer-based language embedding approach for LID tasks by integrating an RNN transducer model into a language embedding framework. Benefiting from the advantages of the RNN transducer's linguistic representation capability, the proposed method can exploit both phonetically-aware acoustic features and explicit linguistic features for LID tasks. Experiments were carried out on the large-scale multilingual LibriSpeech and VoxLingua107 datasets. Experimental results showed the proposed method significantly improves the performance on LID tasks with 12% to 59% and 16% to 24% relative improvement on in-domain and cross-domain datasets, respectively.

Ruiteng Zhang, Jianguo Wei, Xugang Lu et al.

Speaker embedding is an important front-end module to explore discriminative speaker features for many speech applications where speaker information is needed. Current SOTA backbone networks for speaker embedding are designed to aggregate multi-scale features from an utterance with multi-branch network architectures for speaker representation. However, naively adding many branches of multi-scale features with the simple fully convolutional operation could not efficiently improve the performance due to the rapid increase of model parameters and computational complexity. Therefore, in the most current state-of-the-art network architectures, only a few branches corresponding to a limited number of temporal scales could be designed for speaker embeddings. To address this problem, in this paper, we propose an effective temporal multi-scale (TMS) model where multi-scale branches could be efficiently designed in a speaker embedding network almost without increasing computational costs. The new model is based on the conventional TDNN, where the network architecture is smartly separated into two modeling operators: a channel-modeling operator and a temporal multi-branch modeling operator. Adding temporal multi-scale in the temporal multi-branch operator needs only a little bit increase of the number of parameters, and thus save more computational budget for adding more branches with large temporal scales. Moreover, in the inference stage, we further developed a systemic re-parameterization method to convert the TMS-based model into a single-path-based topology in order to increase inference speed. We investigated the performance of the new TMS method for automatic speaker verification (ASV) on in-domain and out-of-domain conditions. Results show that the TMS-based model obtained a significant increase in the performance over the SOTA ASV models, meanwhile, had a faster inference speed.

Peng Shen, Xugang Lu, Hisashi Kawai

For Mandarin end-to-end (E2E) automatic speech recognition (ASR) tasks, compared to character-based modeling units, pronunciation-based modeling units could improve the sharing of modeling units in model training but meet homophone problems. In this study, we propose to use a novel pronunciation-aware unique character encoding for building E2E RNN-T-based Mandarin ASR systems. The proposed encoding is a combination of pronunciation-base syllable and character index (CI). By introducing the CI, the RNN-T model can overcome the homophone problem while utilizing the pronunciation information for extracting modeling units. With the proposed encoding, the model outputs can be converted into the final recognition result through a one-to-one mapping. We conducted experiments on Aishell and MagicData datasets, and the experimental results showed the effectiveness of the proposed method.

Kuo-Hsuan Hung, Kuan-Chen Wang, Kai-Chun Liu et al.

Electrocardiogram (ECG) is an important non-invasive method for diagnosing cardiovascular disease. However, ECG signals are susceptible to noise contamination, such as electrical interference or signal wandering, which reduces diagnostic accuracy. Various ECG denoising methods have been proposed, but most existing methods yield suboptimal performance under very noisy conditions or require several steps during inference, leading to latency during online processing. In this paper, we propose a novel ECG denoising model, namely Mamba-based ECG Enhancer (MECG-E), which leverages the Mamba architecture known for its fast inference and outstanding nonlinear mapping capabilities. Experimental results indicate that MECG-E surpasses several well-known existing models across multiple metrics under different noise conditions. Additionally, MECG-E requires less inference time than state-of-the-art diffusion-based ECG denoisers, demonstrating the model's functionality and efficiency.

Xugang Lu, Peng Shen, Yu Tsao et al.

Transferring linguistic knowledge from a pretrained language model (PLM) to an acoustic model has been shown to greatly improve the performance of automatic speech recognition (ASR). However, due to the heterogeneous feature distributions in cross-modalities, designing an effective model for feature alignment and knowledge transfer between linguistic and acoustic sequences remains a challenging task. Optimal transport (OT), which efficiently measures probability distribution discrepancies, holds great potential for aligning and transferring knowledge between acoustic and linguistic modalities. Nonetheless, the original OT treats acoustic and linguistic feature sequences as two unordered sets in alignment and neglects temporal order information during OT coupling estimation. Consequently, a time-consuming pretraining stage is required to learn a good alignment between the acoustic and linguistic representations. In this paper, we propose a Temporal Order Preserved OT (TOT)-based Cross-modal Alignment and Knowledge Transfer (CAKT) (TOT-CAKT) for ASR. In the TOT-CAKT, local neighboring frames of acoustic sequences are smoothly mapped to neighboring regions of linguistic sequences, preserving their temporal order relationship in feature alignment and matching. With the TOT-CAKT model framework, we conduct Mandarin ASR experiments with a pretrained Chinese PLM for linguistic knowledge transfer. Our results demonstrate that the proposed TOT-CAKT significantly improves ASR performance compared to several state-of-the-art models employing linguistic knowledge transfer, and addresses the weaknesses of the original OT-based method in sequential feature alignment for ASR.

Hao Shi, Yuan Gao, Xugang Lu et al.

Large Language Models (LLMs) are strong decoders for Serialized Output Training (SOT) in two-talker Automatic Speech Recognition (ASR), yet their performance degrades substantially in challenging conditions such as three-talker mixtures. A key limitation is that current systems inject acoustic evidence only through a projected prefix, which can be lossy and imperfectly aligned with the LLM input space, providing insufficient fine-grained grounding during decoding. Addressing this limitation is crucial for robust multi-talker ASR, especially in three-talker mixtures. This paper improves LLM-based multi-talker ASR by explicitly injecting talker-aware acoustic evidence into the decoder. We first revisit Connectionist Temporal Classification (CTC)-derived prefix prompting and compare three variants with increasing acoustic content. The CTC information is obtained using the serialized CTC proposed in our previous works. While acoustic-enriched prompts outperform the SOT-only baseline, prefix-only conditioning remains inadequate for three-talker mixtures. We therefore propose a lightweight gated residual cross-attention adapter and design a two-stage acoustic adaptation framework based on low-rank updates (LoRA). In Stage 1, we insert gated cross-attention adapters after the self-attention sub-layer to stably inject acoustic embeddings as external memory. In Stage 2, we refine both the cross-attention adapters and the pretrained LLM's self-attention projections using parameter-efficient LoRA, improving robustness for large backbones under limited data; the learned updates are merged into the base weights for inference. Experiments on Libri2Mix/Libri3Mix under clean and noisy conditions show consistent gains, with particularly large improvements in three-talker settings.

Meng-Ping Lin, Jen-Cheng Hou, Chia-Wei Chen et al.

Speech enhancement (SE) aims to improve the quality and intelligibility of speech in noisy environments. Recent studies have shown that incorporating visual cues in audio signal processing can enhance SE performance. Given that human speech communication naturally involves audio, visual, and linguistic modalities, it is reasonable to expect additional improvements by integrating linguistic information. However, effectively bridging these modality gaps, particularly during knowledge transfer remains a significant challenge. In this paper, we propose a novel multi-modal learning framework, termed DLAV-SE, which leverages a diffusion-based model integrating audio, visual, and linguistic information for audio-visual speech enhancement (AVSE). Within this framework, the linguistic modality is modeled using a pretrained language model (PLM), which transfers linguistic knowledge to the audio-visual domain through a cross-modal knowledge transfer (CMKT) mechanism during training. After training, the PLM is no longer required at inference, as its knowledge is embedded into the AVSE model through the CMKT process. We conduct a series of SE experiments to evaluate the effectiveness of our approach. Results show that the proposed DLAV-SE system significantly improves speech quality and reduces generative artifacts, such as phonetic confusion, compared to state-of-the-art (SOTA) methods. Furthermore, visualization analyses confirm that the CMKT method enhances the generation quality of the AVSE outputs. These findings highlight both the promise of diffusion-based methods for advancing AVSE and the value of incorporating linguistic information to further improve system performance.

Peng Shen, Xugang Lu, Hisashi Kawai

Multi-talker overlapped speech recognition remains a significant challenge, requiring not only speech recognition but also speaker diarization tasks to be addressed. In this paper, to better address these tasks, we first introduce speaker labels into an autoregressive transformer-based speech recognition model to support multi-speaker overlapped speech recognition. Then, to improve speaker diarization, we propose a novel speaker mask branch to detection the speech segments of individual speakers. With the proposed model, we can perform both speech recognition and speaker diarization tasks simultaneously using a single model. Experimental results on the LibriSpeech-based overlapped dataset demonstrate the effectiveness of the proposed method in both speech recognition and speaker diarization tasks, particularly enhancing the accuracy of speaker diarization in relatively complex multi-talker scenarios.

Kuo-Hsuan Hung, Xugang Lu, Szu-Wei Fu et al.

Linguistic knowledge plays a crucial role in spoken language comprehension. It provides essential semantic and syntactic context for speech perception in noisy environments. However, most speech enhancement (SE) methods predominantly rely on acoustic features to learn the mapping relationship between noisy and clean speech, with limited exploration of linguistic integration. While text-informed SE approaches have been investigated, they often require explicit speech-text alignment or externally provided textual data, constraining their practicality in real-world scenarios. Additionally, using text as input poses challenges in aligning linguistic and acoustic representations due to their inherent differences. In this study, we propose the Cross-Modality Knowledge Transfer (CMKT) learning framework, which leverages pre-trained large language models (LLMs) to infuse linguistic knowledge into SE models without requiring text input or LLMs during inference. Furthermore, we introduce a misalignment strategy to improve knowledge transfer. This strategy applies controlled temporal shifts, encouraging the model to learn more robust representations. Experimental evaluations demonstrate that CMKT consistently outperforms baseline models across various SE architectures and LLM embeddings, highlighting its adaptability to different configurations. Additionally, results on Mandarin and English datasets confirm its effectiveness across diverse linguistic conditions, further validating its robustness. Moreover, CMKT remains effective even in scenarios without textual data, underscoring its practicality for real-world applications. By bridging the gap between linguistic and acoustic modalities, CMKT offers a scalable and innovative solution for integrating linguistic knowledge into SE models, leading to substantial improvements in both intelligibility and enhancement performance.

Peng Shen, Xugang Lu, Hisashi Kawai

Speech recognition systems often face challenges due to domain mismatch, particularly in real-world applications where domain-specific data is unavailable because of data accessibility and confidentiality constraints. Inspired by Retrieval-Augmented Generation (RAG) techniques for large language models (LLMs), this paper introduces a LLM-based retrieval-augmented speech recognition method that incorporates domain-specific textual data at the inference stage to enhance recognition performance. Rather than relying on domain-specific textual data during the training phase, our model is trained to learn how to utilize textual information provided in prompts for LLM decoder to improve speech recognition performance. Benefiting from the advantages of the RAG retrieval mechanism, our approach efficiently accesses locally available domain-specific documents, ensuring a convenient and effective process for solving domain mismatch problems. Experiments conducted on the CSJ database demonstrate that the proposed method significantly improves speech recognition accuracy and achieves state-of-the-art results on the CSJ dataset, even without relying on the full training data.

Cho-Yuan Lee, Kuan-Chen Wang, Kai-Chun Liu et al.

In practical scenarios involving the measurement of surface electromyography (sEMG) in muscles, particularly those areas near the heart, one of the primary sources of contamination is the presence of electrocardiogram (ECG) signals. To assess the quality of real-world sEMG data more effectively, this study proposes QASE-net, a new non-intrusive model that predicts the SNR of sEMG signals. QASE-net combines CNN-BLSTM with attention mechanisms and follows an end-to-end training strategy. Our experimental framework utilizes real-world sEMG and ECG data from two open-access databases, the Non-Invasive Adaptive Prosthetics Database and the MIT-BIH Normal Sinus Rhythm Database, respectively. The experimental results demonstrate the superiority of QASE-net over the previous assessment model, exhibiting significantly reduced prediction errors and notably higher linear correlations with the ground truth. These findings show the potential of QASE-net to substantially enhance the reliability and precision of sEMG quality assessment in practical applications.

Xugang Lu, Peng Shen, Yu Tsao et al.

Aligning acoustic and linguistic representations is a central challenge to bridge the pre-trained models in knowledge transfer for automatic speech recognition (ASR). This alignment is inherently structured and asymmetric: while multiple consecutive acoustic frames typically correspond to a single linguistic token (many-to-one), certain acoustic transition regions may relate to multiple adjacent tokens (one-to-many). Moreover, acoustic sequences often include frames with no linguistic counterpart, such as background noise or silence may lead to imbalanced matching conditions. In this work, we take a new insight to regard alignment and matching as a detection problem, where the goal is to identify meaningful correspondences with high precision and recall ensuring full coverage of linguistic tokens while flexibly handling redundant or noisy acoustic frames in transferring linguistic knowledge for ASR. Based on this new insight, we propose an unbalanced optimal transport-based alignment model that explicitly handles distributional mismatch and structural asymmetries with soft and partial matching between acoustic and linguistic modalities. Our method ensures that every linguistic token is grounded in at least one acoustic observation, while allowing for flexible, probabilistic mappings from acoustic to linguistic units. We evaluate our proposed model with experiments on an CTC-based ASR system with a pre-trained language model for knowledge transfer. Experimental results demonstrate the effectiveness of our approach in flexibly controlling degree of matching and hence to improve ASR performance.

Xugang Lu, Peng Shen, Yu Tsao et al.

Transferring linguistic knowledge from a pretrained language model (PLM) to acoustic feature learning has proven effective in enhancing end-to-end automatic speech recognition (E2E-ASR). However, aligning representations between linguistic and acoustic modalities remains a challenge due to inherent modality gaps. Optimal transport (OT) has shown promise in mitigating these gaps by minimizing the Wasserstein distance (WD) between linguistic and acoustic feature distributions. However, previous OT-based methods overlook structural relationships, treating feature vectors as unordered sets. To address this, we propose Graph Matching Optimal Transport (GM-OT), which models linguistic and acoustic sequences as structured graphs. Nodes represent feature embeddings, while edges capture temporal and sequential relationships. GM-OT minimizes both WD (between nodes) and Gromov-Wasserstein distance (GWD) (between edges), leading to a fused Gromov-Wasserstein distance (FGWD) formulation. This enables structured alignment and more efficient knowledge transfer compared to existing OT-based approaches. Theoretical analysis further shows that prior OT-based methods in linguistic knowledge transfer can be viewed as a special case within our GM-OT framework. We evaluate GM-OT on Mandarin ASR using a CTC-based E2E-ASR system with a PLM for knowledge transfer. Experimental results demonstrate significant performance gains over state-of-the-art models, validating the effectiveness of our approach.

Wenhao Yang, Jianguo Wei, Wenhuan Lu et al.

Recent research in speaker verification has increasingly focused on achieving robust and reliable recognition under challenging channel conditions and noisy environments. Identifying speakers in radio communications is particularly difficult due to inherent limitations such as constrained bandwidth and pervasive noise interference. To address this issue, we present a Channel Robust Speaker Learning (CRSL) framework that enhances the robustness of the current speaker verification pipeline, considering data source, data augmentation, and the efficiency of model transfer processes. Our framework introduces an augmentation module that mitigates bandwidth variations in radio speech datasets by manipulating the bandwidth of training inputs. It also addresses unknown noise by introducing noise within the manifold space. Additionally, we propose an efficient fine-tuning method that reduces the need for extensive additional training time and large amounts of data. Moreover, we develop a toolkit for assembling a large-scale radio speech corpus and establish a benchmark specifically tailored for radio scenario speaker verification studies. Experimental results demonstrate that our proposed methodology effectively enhances performance and mitigates degradation caused by radio transmission in speaker verification tasks. The code will be available on Github.

Rong Chao, Cheng Yu, Szu-Wei Fu et al.

Speech enhancement (SE) performance has improved considerably owing to the use of deep learning models as a base function. Herein, we propose a perceptual contrast stretching (PCS) approach to further improve SE performance. The PCS is derived based on the critical band importance function and is applied to modify the targets of the SE model. Specifically, the contrast of target features is stretched based on perceptual importance, thereby improving the overall SE performance. Compared with post-processing-based implementations, incorporating PCS into the training phase preserves performance and reduces online computation. Notably, PCS can be combined with different SE model architectures and training criteria. Furthermore, PCS does not affect the causality or convergence of SE model training. Experimental results on the VoiceBank-DEMAND dataset show that the proposed method can achieve state-of-the-art performance on both causal (PESQ score = 3.07) and noncausal (PESQ score = 3.35) SE tasks.

Xugang Lu, Peng Shen, Yu Tsao et al.

In order to reduce domain discrepancy to improve the performance of cross-domain spoken language identification (SLID) system, as an unsupervised domain adaptation (UDA) method, we have proposed a joint distribution alignment (JDA) model based on optimal transport (OT). A discrepancy measurement based on OT was adopted for JDA between training and test data sets. In our previous study, it was supposed that the training and test sets share the same label space. However, in real applications, the label space of the test set is only a subset of that of the training set. Fully matching training and test domains for distribution alignment may introduce negative domain transfer. In this paper, we propose an JDA model based on partial optimal transport (POT), i.e., only partial couplings of OT are allowed during JDA. Moreover, since the label of test data is unknown, in the POT, a soft weighting on the coupling based on transport cost is adaptively set during domain alignment. Experiments were carried out on a cross-domain SLID task to evaluate the proposed UDA. Results showed that our proposed UDA significantly improved the performance due to the consideration of the partial couplings in OT.

Tassadaq Hussain, Wei-Chien Wang, Mandar Gogate et al.

In acoustic signal processing, the target signals usually carry semantic information, which is encoded in a hierarchal structure of short and long-term contexts. However, the background noise distorts these structures in a nonuniform way. The existing deep acoustic signal enhancement (ASE) architectures ignore this kind of local and global effect. To address this problem, we propose to integrate a novel temporal attentive-pooling (TAP) mechanism into a conventional convolutional recurrent neural network, termed as TAP-CRNN. The proposed approach considers both global and local attention for ASE tasks. Specifically, we first utilize a convolutional layer to extract local information of the acoustic signals and then a recurrent neural network (RNN) architecture is used to characterize temporal contextual information. Second, we exploit a novelattention mechanism to contextually process salient regions of the noisy signals. The proposed ASE system is evaluated using a benchmark infant cry dataset and compared with several well-known methods. It is shown that the TAPCRNN can more effectively reduce noise components from infant cry signals in unseen background noises at challenging signal-to-noise levels.

Hsin-Yi Lin, Huan-Hsin Tseng, Xugang Lu et al.

This paper presents a novel discriminator-constrained optimal transport network (DOTN) that performs unsupervised domain adaptation for speech enhancement (SE), which is an essential regression task in speech processing. The DOTN aims to estimate clean references of noisy speech in a target domain, by exploiting the knowledge available from the source domain. The domain shift between training and testing data has been reported to be an obstacle to learning problems in diverse fields. Although rich literature exists on unsupervised domain adaptation for classification, the methods proposed, especially in regressions, remain scarce and often depend on additional information regarding the input data. The proposed DOTN approach tactically fuses the optimal transport (OT) theory from mathematical analysis with generative adversarial frameworks, to help evaluate continuous labels in the target domain. The experimental results on two SE tasks demonstrate that by extending the classical OT formulation, our proposed DOTN outperforms previous adversarial domain adaptation frameworks in a purely unsupervised manner.

Ruiteng Zhang, Jianguo Wei, Wenhuan Lu et al.

Automatic speaker verification (ASV) systems, which determine whether two speeches are from the same speaker, mainly focus on verification accuracy while ignoring inference speed. However, in real applications, both inference speed and verification accuracy are essential. This study proposes cross-sequential re-parameterization (CS-Rep), a novel topology re-parameterization strategy for multi-type networks, to increase the inference speed and verification accuracy of models. CS-Rep solves the problem that existing re-parameterization methods are unsuitable for typical ASV backbones. When a model applies CS-Rep, the training-period network utilizes a multi-branch topology to capture speaker information, whereas the inference-period model converts to a time-delay neural network (TDNN)-like plain backbone with stacked TDNN layers to achieve the fast inference speed. Based on CS-Rep, an improved TDNN with friendly test and deployment called Rep-TDNN is proposed. Compared with the state-of-the-art model ECAPA-TDNN, which is highly recognized in the industry, Rep-TDNN increases the actual inference speed by about 50% and reduces the EER by 10%. The code will be released.

Szu-Wei Fu, Cheng Yu, Tsun-An Hsieh et al.

The discrepancy between the cost function used for training a speech enhancement model and human auditory perception usually makes the quality of enhanced speech unsatisfactory. Objective evaluation metrics which consider human perception can hence serve as a bridge to reduce the gap. Our previously proposed MetricGAN was designed to optimize objective metrics by connecting the metric with a discriminator. Because only the scores of the target evaluation functions are needed during training, the metrics can even be non-differentiable. In this study, we propose a MetricGAN+ in which three training techniques incorporating domain-knowledge of speech processing are proposed. With these techniques, experimental results on the VoiceBank-DEMAND dataset show that MetricGAN+ can increase PESQ score by 0.3 compared to the previous MetricGAN and achieve state-of-the-art results (PESQ score = 3.15).

Xugang Lu, Peng Shen, Yu Tsao et al.

Generative probability models are widely used for speaker verification (SV). However, the generative models are lack of discriminative feature selection ability. As a hypothesis test, the SV can be regarded as a binary classification task which can be designed as a Siamese neural network (SiamNN) with discriminative training. However, in most of the discriminative training for SiamNN, only the distribution of pair-wised sample distances is considered, and the additional discriminative information in joint distribution of samples is ignored. In this paper, we propose a novel SiamNN with consideration of the joint distribution of samples. The joint distribution of samples is first formulated based on a joint Bayesian (JB) based generative model, then a SiamNN is designed with dense layers to approximate the factorized affine transforms as used in the JB model. By initializing the SiamNN with the learned model parameters of the JB model, we further train the model parameters with the pair-wised samples as a binary discrimination task for SV. We carried out SV experiments on data corpus of speakers in the wild (SITW) and VoxCeleb. Experimental results showed that our proposed model improved the performance with a large margin compared with state of the art models for SV.

Yu-Wen Chen, Kuo-Hsuan Hung, Shang-Yi Chuang et al.

Synthesized speech from articulatory movements can have real-world use for patients with vocal cord disorders, situations requiring silent speech, or in high-noise environments. In this work, we present EMA2S, an end-to-end multimodal articulatory-to-speech system that directly converts articulatory movements to speech signals. We use a neural-network-based vocoder combined with multimodal joint-training, incorporating spectrogram, mel-spectrogram, and deep features. The experimental results confirm that the multimodal approach of EMA2S outperforms the baseline system in terms of both objective evaluation and subjective evaluation metrics. Moreover, results demonstrate that joint mel-spectrogram and deep feature loss training can effectively improve system performance.

Xugang Lu, Peng Shen, Yu Tsao et al.

The speaker verification (SV) task is to decide whether an utterance is spoken by a target or an imposter speaker. For most studies, a log-likelihood ratio (LLR) score is estimated based on a generative probability model on speaker features and compared with a threshold for making a decision. However, the generative model usually focuses on individual feature distributions, does not have the discriminative feature selection ability, and is easy to be distracted by nuisance features. The SV could be formulated as a binary discrimination task where neural network-based discriminative learning could be applied. In discriminative learning, the nuisance features could be removed with the help of label supervision. However, discriminative learning pays more attention to classification boundaries and is prone to overfitting to a training set which may result in bad generalization on a test set. Thus, we propose a hybrid learning framework, i.e., coupling a joint Bayesian (JB) generative model structure and parameters with a neural discriminative learning framework for SV. A two-branch Siamese neural network is built with dense layers that are coupled with factorized affine transforms as used in the JB model. The LLR score estimation in the JB model is formulated according to the distance metric in the discriminative learning framework. By initializing the two-branch neural network with the generatively learned model parameters of the JB model, we train the model parameters with the pairwise samples as a binary discrimination task. Moreover, a direct evaluation metric in SV based on minimum empirical Bayes risk is designed and integrated as an objective function in discriminative learning. We carried out SV experiments on Speakers in the wild and Voxceleb. Experimental results showed that our proposed model improved the performance with a large margin compared with state-of-art models for SV.

Xugang Lu, Peng Shen, Yu Tsao et al.

Due to the mismatch of statistical distributions of acoustic speech between training and testing sets, the performance of spoken language identification (SLID) could be drastically degraded. In this paper, we propose an unsupervised neural adaptation model to deal with the distribution mismatch problem for SLID. In our model, we explicitly formulate the adaptation as to reduce the distribution discrepancy on both feature and classifier for training and testing data sets. Moreover, inspired by the strong power of the optimal transport (OT) to measure distribution discrepancy, a Wasserstein distance metric is designed in the adaptation loss. By minimizing the classification loss on the training data set with the adaptation loss on both training and testing data sets, the statistical distribution difference between training and testing domains is reduced. We carried out SLID experiments on the oriental language recognition (OLR) challenge data corpus where the training and testing data sets were collected from different conditions. Our results showed that significant improvements were achieved on the cross domain test tasks.

Tsun-An Hsieh, Cheng Yu, Szu-Wei Fu et al.

Speech enhancement (SE) aims to improve speech quality and intelligibility, which are both related to a smooth transition in speech segments that may carry linguistic information, e.g. phones and syllables. In this study, we propose a novel phone-fortified perceptual loss (PFPL) that takes phonetic information into account for training SE models. To effectively incorporate the phonetic information, the PFPL is computed based on latent representations of the wav2vec model, a powerful self-supervised encoder that renders rich phonetic information. To more accurately measure the distribution distances of the latent representations, the PFPL adopts the Wasserstein distance as the distance measure. Our experimental results first reveal that the PFPL is more correlated with the perceptual evaluation metrics, as compared to signal-level losses. Moreover, the results showed that the PFPL can enable a deep complex U-Net SE model to achieve highly competitive performance in terms of standardized quality and intelligibility evaluations on the Voice Bank-DEMAND dataset.

Yen-Ju Lu, Chien-Feng Liao, Xugang Lu et al.

In noisy conditions, knowing speech contents facilitates listeners to more effectively suppress background noise components and to retrieve pure speech signals. Previous studies have also confirmed the benefits of incorporating phonetic information in a speech enhancement (SE) system to achieve better denoising performance. To obtain the phonetic information, we usually prepare a phoneme-based acoustic model, which is trained using speech waveforms and phoneme labels. Despite performing well in normal noisy conditions, when operating in very noisy conditions, however, the recognized phonemes may be erroneous and thus misguide the SE process. To overcome the limitation, this study proposes to incorporate the broad phonetic class (BPC) information into the SE process. We have investigated three criteria to build the BPC, including two knowledge-based criteria: place and manner of articulatory and one data-driven criterion. Moreover, the recognition accuracies of BPCs are much higher than that of phonemes, thus providing more accurate phonetic information to guide the SE process under very noisy conditions. Experimental results demonstrate that the proposed SE with the BPC information framework can achieve notable performance improvements over the baseline system and an SE system using monophonic information in terms of both speech quality intelligibility on the TIMIT dataset.

Tsun-An Hsieh, Hsin-Min Wang, Xugang Lu et al.

Due to the simple design pipeline, end-to-end (E2E) neural models for speech enhancement (SE) have attracted great interest. In order to improve the performance of the E2E model, the locality and temporal sequential properties of speech should be efficiently taken into account when modelling. However, in most current E2E models for SE, these properties are either not fully considered or are too complex to be realized. In this paper, we propose an efficient E2E SE model, termed WaveCRN. In WaveCRN, the speech locality feature is captured by a convolutional neural network (CNN), while the temporal sequential property of the locality feature is modeled by stacked simple recurrent units (SRU). Unlike a conventional temporal sequential model that uses a long short-term memory (LSTM) network, which is difficult to parallelize, SRU can be efficiently parallelized in calculation with even fewer model parameters. In addition, in order to more effectively suppress the noise components in the input noisy speech, we derive a novel restricted feature masking (RFM) approach that performs enhancement on the feature maps in the hidden layers; this is different from the approach that applies the estimated ratio mask on the noisy spectral features, which is commonly used in speech separation methods. Experimental results on speech denoising and compressed speech restoration tasks confirm that with the lightweight architecture of SRU and the feature-mapping-based RFM, WaveCRN performs comparably with other state-of-the-art approaches with notably reduced model complexity and inference time.

Haipeng Sun, Rui Wang, Kehai Chen et al.

Unsupervised neural machine translation (UNMT) has recently attracted great interest in the machine translation community. The main advantage of the UNMT lies in its easy collection of required large training text sentences while with only a slightly worse performance than supervised neural machine translation which requires expensive annotated translation pairs on some translation tasks. In most studies, the UMNT is trained with clean data without considering its robustness to the noisy data. However, in real-world scenarios, there usually exists noise in the collected input sentences which degrades the performance of the translation system since the UNMT is sensitive to the small perturbations of the input sentences. In this paper, we first time explicitly take the noisy data into consideration to improve the robustness of the UNMT based systems. First of all, we clearly defined two types of noises in training sentences, i.e., word noise and word order noise, and empirically investigate its effect in the UNMT, then we propose adversarial training methods with denoising process in the UNMT. Experimental results on several language pairs show that our proposed methods substantially improved the robustness of the conventional UNMT systems in noisy scenarios.

Cheng Yu, Ryandhimas E. Zezario, Syu-Siang Wang et al.

Deep learning-based models have greatly advanced the performance of speech enhancement (SE) systems. However, two problems remain unsolved, which are closely related to model generalizability to noisy conditions: (1) mismatched noisy condition during testing, i.e., the performance is generally sub-optimal when models are tested with unseen noise types that are not involved in the training data; (2) local focus on specific noisy conditions, i.e., models trained using multiple types of noises cannot optimally remove a specific noise type even though the noise type has been involved in the training data. These problems are common in real applications. In this paper, we propose a novel denoising autoencoder with a multi-branched encoder (termed DAEME) model to deal with these two problems. In the DAEME model, two stages are involved: training and testing. In the training stage, we build multiple component models to form a multi-branched encoder based on a decision tree (DSDT). The DSDT is built based on prior knowledge of speech and noisy conditions (the speaker, environment, and signal factors are considered in this paper), where each component of the multi-branched encoder performs a particular mapping from noisy to clean speech along the branch in the DSDT. Finally, a decoder is trained on top of the multi-branched encoder. In the testing stage, noisy speech is first processed by each component model. The multiple outputs from these models are then integrated into the decoder to determine the final enhanced speech. Experimental results show that DAEME is superior to several baseline models in terms of objective evaluation metrics, automatic speech recognition results, and quality in subjective human listening tests.

Xugang Lu, Peng Shen, Sheng Li et al.

A major advantage of a deep convolutional neural network (CNN) is that the focused receptive field size is increased by stacking multiple convolutional layers. Accordingly, the model can explore the long-range dependency of features from the top layers. However, a potential limitation of the network is that the discriminative features from the bottom layers (which can model the short-range dependency) are smoothed out in the final representation. This limitation is especially evident in the acoustic event classification (AEC) task, where both short- and long-duration events are involved in an audio clip and needed to be classified. In this paper, we propose a cross-scale attention (CSA) model, which explicitly integrates features from different scales to form the final representation. Moreover, we propose the adoption of the attention mechanism to specify the weights of local and global features based on the spatial and temporal characteristics of acoustic events. Using mathematic formulations, we further reveal that the proposed CSA model can be regarded as a weighted residual CNN (ResCNN) model when the ResCNN is used as a backbone model. We tested the proposed model on two AEC datasets: one is an urban AEC task, and the other is an AEC task in smart car environments. Experimental results show that the proposed CSA model can effectively improve the performance of current state-of-the-art deep learning algorithms.

Chien-Feng Liao, Yu Tsao, Xugang Lu et al.

In a noisy environment, a lossy speech signal can be automatically restored by a listener if he/she knows the language well. That is, with the built-in knowledge of a "language model", a listener may effectively suppress noise interference and retrieve the target speech signals. Accordingly, we argue that familiarity with the underlying linguistic content of spoken utterances benefits speech enhancement (SE) in noisy environments. In this study, in addition to the conventional modeling for learning the acoustic noisy-clean speech mapping, an abstract symbolic sequential modeling is incorporated into the SE framework. This symbolic sequential modeling can be regarded as a "linguistic constraint" in learning the acoustic noisy-clean speech mapping function. In this study, the symbolic sequences for acoustic signals are obtained as discrete representations with a Vector Quantized Variational Autoencoder algorithm. The obtained symbols are able to capture high-level phoneme-like content from speech signals. The experimental results demonstrate that the proposed framework can obtain notable performance improvement in terms of perceptual evaluation of speech quality (PESQ) and short-time objective intelligibility (STOI) on the TIMIT dataset.

Wei-Jen Lee, Syu-Siang Wang, Fei Chen et al.

Reverberation, which is generally caused by sound reflections from walls, ceilings, and floors, can result in severe performance degradation of acoustic applications. Due to a complicated combination of attenuation and time-delay effects, the reverberation property is difficult to characterize, and it remains a challenging task to effectively retrieve the anechoic speech signals from reverberation ones. In the present study, we proposed a novel integrated deep and ensemble learning algorithm (IDEA) for speech dereverberation. The IDEA consists of offline and online phases. In the offline phase, we train multiple dereverberation models, each aiming to precisely dereverb speech signals in a particular acoustic environment; then a unified fusion function is estimated that aims to integrate the information of multiple dereverberation models. In the online phase, an input utterance is first processed by each of the dereverberation models. The outputs of all models are integrated accordingly to generate the final anechoic signal. We evaluated the IDEA on designed acoustic environments, including both matched and mismatched conditions of the training and testing data. Experimental results confirm that the proposed IDEA outperforms single deep-neural-network-based dereverberation model with the same model architecture and training data.

Szu-Wei Fu, Tao-Wei Wang, Yu Tsao et al.

Speech enhancement model is used to map a noisy speech to a clean speech. In the training stage, an objective function is often adopted to optimize the model parameters. However, in most studies, there is an inconsistency between the model optimization criterion and the evaluation criterion on the enhanced speech. For example, in measuring speech intelligibility, most of the evaluation metric is based on a short-time objective intelligibility (STOI) measure, while the frame based minimum mean square error (MMSE) between estimated and clean speech is widely used in optimizing the model. Due to the inconsistency, there is no guarantee that the trained model can provide optimal performance in applications. In this study, we propose an end-to-end utterance-based speech enhancement framework using fully convolutional neural networks (FCN) to reduce the gap between the model optimization and evaluation criterion. Because of the utterance-based optimization, temporal correlation information of long speech segments, or even at the entire utterance level, can be considered when perception-based objective functions are used for the direct optimization. As an example, we implement the proposed FCN enhancement framework to optimize the STOI measure. Experimental results show that the STOI of test speech is better than conventional MMSE-optimized speech due to the consistency between the training and evaluation target. Moreover, by integrating the STOI in model optimization, the intelligibility of human subjects and automatic speech recognition (ASR) system on the enhanced speech is also substantially improved compared to those generated by the MMSE criterion.

Szu-Wei Fu, Ting-yao Hu, Yu Tsao et al.

This paper aims to address two issues existing in the current speech enhancement methods: 1) the difficulty of phase estimations; 2) a single objective function cannot consider multiple metrics simultaneously. To solve the first problem, we propose a novel convolutional neural network (CNN) model for complex spectrogram enhancement, namely estimating clean real and imaginary (RI) spectrograms from noisy ones. The reconstructed RI spectrograms are directly used to synthesize enhanced speech waveforms. In addition, since log-power spectrogram (LPS) can be represented as a function of RI spectrograms, its reconstruction is also considered as another target. Thus a unified objective function, which combines these two targets (reconstruction of RI spectrograms and LPS), is equivalent to simultaneously optimizing two commonly used objective metrics: segmental signal-to-noise ratio (SSNR) and logspectral distortion (LSD). Therefore, the learning process is called multi-metrics learning (MML). Experimental results confirm the effectiveness of the proposed CNN with RI spectrograms and MML in terms of improved standardized evaluation metrics on a speech enhancement task.

Szu-Wei Fu, Yu Tsao, Xugang Lu et al.

This study proposes a fully convolutional network (FCN) model for raw waveform-based speech enhancement. The proposed system performs speech enhancement in an end-to-end (i.e., waveform-in and waveform-out) manner, which dif-fers from most existing denoising methods that process the magnitude spectrum (e.g., log power spectrum (LPS)) only. Because the fully connected layers, which are involved in deep neural networks (DNN) and convolutional neural networks (CNN), may not accurately characterize the local information of speech signals, particularly with high frequency components, we employed fully convolutional layers to model the waveform. More specifically, FCN consists of only convolutional layers and thus the local temporal structures of speech signals can be efficiently and effectively preserved with relatively few weights. Experimental results show that DNN- and CNN-based models have limited capability to restore high frequency components of waveforms, thus leading to decreased intelligibility of enhanced speech. By contrast, the proposed FCN model can not only effectively recover the waveforms but also outperform the LPS-based DNN baseline in terms of short-time objective intelligibility (STOI) and perceptual evaluation of speech quality (PESQ). In addition, the number of model parameters in FCN is approximately only 0.2% compared with that in both DNN and CNN.

Syu-Siang Wang, Alan Chern, Yu Tsao et al.

For most of the state-of-the-art speech enhancement techniques, a spectrogram is usually preferred than the respective time-domain raw data since it reveals more compact presentation together with conspicuous temporal information over a long time span. However, the short-time Fourier transform (STFT) that creates the spectrogram in general distorts the original signal and thereby limits the capability of the associated speech enhancement techniques. In this study, we propose a novel speech enhancement method that adopts the algorithms of discrete wavelet packet transform (DWPT) and nonnegative matrix factorization (NMF) in order to conquer the aforementioned limitation. In brief, the DWPT is first applied to split a time-domain speech signal into a series of subband signals without introducing any distortion. Then we exploit NMF to highlight the speech component for each subband. Finally, the enhanced subband signals are joined together via the inverse DWPT to reconstruct a noise-reduced signal in time domain. We evaluate the proposed DWPT-NMF based speech enhancement method on the MHINT task. Experimental results show that this new method behaves very well in prompting speech quality and intelligibility and it outperforms the convnenitional STFT-NMF based method.