SAV-SE: Scene-aware Audio-Visual Speech Enhancement with Selective State Space Model

Speech enhancement plays an essential role in various applications, and the integration of visual information has been demonstrated to bring substantial advantages. However, the majority of current research concentrates on the examination of facial and lip movements, which can be compromised or entirely inaccessible in scenarios where occlusions occur or when the camera view is distant. Whereas contextual visual cues from the surrounding environment have been overlooked: for example, when we see a dog bark, our brain has the innate ability to discern and filter out the barking noise. To this end, in this paper, we introduce a novel task, i.e. SAV-SE. To our best knowledge, this is the first proposal to use rich contextual information from synchronized video as auxiliary cues to indicate the type of noise, which eventually improves the speech enhancement performance. Specifically, we propose the VC-S$^2$E method, which incorporates the Conformer and Mamba modules for their complementary strengths. Extensive experiments are conducted on public MUSIC, AVSpeech and AudioSet datasets, where the results demonstrate the superiority of VC-S$^2$E over other competitive methods. We will make the source code publicly available. Project demo page: https://AVSEPage.github.io/

Selective State Space Model for Monaural Speech Enhancement

Voice user interfaces (VUIs) have facilitated the efficient interactions between humans and machines through spoken commands. Since real-word acoustic scenes are complex, speech enhancement plays a critical role for robust VUI. Transformer and its variants, such as Conformer, have demonstrated cutting-edge results in speech enhancement. However, both of them suffers from the quadratic computational complexity with respect to the sequence length, which hampers their ability to handle long sequences. Recently a novel State Space Model called Mamba, which shows strong capability to handle long sequences with linear complexity, offers a solution to address this challenge. In this paper, we propose a novel hybrid convolution-Mamba backbone, denoted as MambaDC, for speech enhancement. Our MambaDC marries the benefits of convolutional networks to model the local interactions and Mamba's ability for modeling long-range global dependencies. We conduct comprehensive experiments within both basic and state-of-the-art (SoTA) speech enhancement frameworks, on two commonly used training targets. The results demonstrate that MambaDC outperforms Transformer, Conformer, and the standard Mamba across all training targets. Built upon the current advanced framework, the use of MambaDC backbone showcases superior results compared to existing \textcolor{black}{SoTA} systems. This sets the stage for efficient long-range global modeling in speech enhancement.

Binaural Selective Attention Model for Target Speaker Extraction

The remarkable ability of humans to selectively focus on a target speaker in cocktail party scenarios is facilitated by binaural audio processing. In this paper, we present a binaural time-domain Target Speaker Extraction model based on the Filter-and-Sum Network (FaSNet). Inspired by human selective hearing, our proposed model introduces target speaker embedding into separators using a multi-head attention-based selective attention block. We also compared two binaural interaction approaches -- the cosine similarity of time-domain signals and inter-channel correlation in learned spectral representations. Our experimental results show that our proposed model outperforms monaural configurations and state-of-the-art multi-channel target speaker extraction models, achieving best-in-class performance with 18.52 dB SI-SDR, 19.12 dB SDR, and 3.05 PESQ scores under anechoic two-speaker test configurations.

An Exploration of Length Generalization in Transformer-Based Speech Enhancement

The use of Transformer architectures has facilitated remarkable progress in speech enhancement. Training Transformers using substantially long speech utterances is often infeasible as self-attention suffers from quadratic complexity. It is a critical and unexplored challenge for a Transformer-based speech enhancement model to learn from short speech utterances and generalize to longer ones. In this paper, we conduct comprehensive experiments to explore the length generalization problem in speech enhancement with Transformer. Our findings first establish that position embedding provides an effective instrument to alleviate the impact of utterance length on Transformer-based speech enhancement. Specifically, we explore four different position embedding schemes to enable length generalization. The results confirm the superiority of relative position embeddings (RPEs) over absolute PE (APEs) in length generalization.

Mamba in Speech: Towards an Alternative to Self-Attention

Transformer and its derivatives have achieved success in diverse tasks across computer vision, natural language processing, and speech processing. To reduce the complexity of computations within the multi-head self-attention mechanism in Transformer, Selective State Space Models (i.e., Mamba) were proposed as an alternative. Mamba exhibited its effectiveness in natural language processing and computer vision tasks, but its superiority has rarely been investigated in speech signal processing. This paper explores solutions for applying Mamba to speech processing using two typical speech processing tasks: speech recognition, which requires semantic and sequential information, and speech enhancement, which focuses primarily on sequential patterns. The results exhibit the superiority of bidirectional Mamba (BiMamba) for speech processing to vanilla Mamba. Moreover, experiments demonstrate the effectiveness of BiMamba as an alternative to the self-attention module in Transformer and its derivates, particularly for the semantic-aware task. The crucial technologies for transferring Mamba to speech are then summarized in ablation studies and the discussion section to offer insights for future research.

When LLMs Meets Acoustic Landmarks: An Efficient Approach to Integrate Speech into Large Language Models for Depression Detection

Depression is a critical concern in global mental health, prompting extensive research into AI-based detection methods. Among various AI technologies, Large Language Models (LLMs) stand out for their versatility in mental healthcare applications. However, their primary limitation arises from their exclusive dependence on textual input, which constrains their overall capabilities. Furthermore, the utilization of LLMs in identifying and analyzing depressive states is still relatively untapped. In this paper, we present an innovative approach to integrating acoustic speech information into the LLMs framework for multimodal depression detection. We investigate an efficient method for depression detection by integrating speech signals into LLMs utilizing Acoustic Landmarks. By incorporating acoustic landmarks, which are specific to the pronunciation of spoken words, our method adds critical dimensions to text transcripts. This integration also provides insights into the unique speech patterns of individuals, revealing the potential mental states of individuals. Evaluations of the proposed approach on the DAIC-WOZ dataset reveal state-of-the-art results when compared with existing Audio-Text baselines. In addition, this approach is not only valuable for the detection of depression but also represents a new perspective in enhancing the ability of LLMs to comprehend and process speech signals.

Speaking in Wavelet Domain: A Simple and Efficient Approach to Speed up Speech Diffusion Model

Recently, Denoising Diffusion Probabilistic Models (DDPMs) have attained leading performances across a diverse range of generative tasks. However, in the field of speech synthesis, although DDPMs exhibit impressive performance, their long training duration and substantial inference costs hinder practical deployment. Existing approaches primarily focus on enhancing inference speed, while approaches to accelerate training a key factor in the costs associated with adding or customizing voices often necessitate complex modifications to the model, compromising their universal applicability. To address the aforementioned challenges, we propose an inquiry: is it possible to enhance the training/inference speed and performance of DDPMs by modifying the speech signal itself? In this paper, we double the training and inference speed of Speech DDPMs by simply redirecting the generative target to the wavelet domain. This method not only achieves comparable or superior performance to the original model in speech synthesis tasks but also demonstrates its versatility. By investigating and utilizing different wavelet bases, our approach proves effective not just in speech synthesis, but also in speech enhancement.

An Empirical Study on the Impact of Positional Encoding in Transformer-based Monaural Speech Enhancement

Transformer architecture has enabled recent progress in speech enhancement. Since Transformers are position-agostic, positional encoding is the de facto standard component used to enable Transformers to distinguish the order of elements in a sequence. However, it remains unclear how positional encoding exactly impacts speech enhancement based on Transformer architectures. In this paper, we perform a comprehensive empirical study evaluating five positional encoding methods, i.e., Sinusoidal and learned absolute position embedding (APE), T5-RPE, KERPLE, as well as the Transformer without positional encoding (No-Pos), across both causal and noncausal configurations. We conduct extensive speech enhancement experiments, involving spectral mapping and masking methods. Our findings establish that positional encoding is not quite helpful for the models in a causal configuration, which indicates that causal attention may implicitly incorporate position information. In a noncausal configuration, the models significantly benefit from the use of positional encoding. In addition, we find that among the four position embeddings, relative position embeddings outperform APEs.

EEG-Derived Voice Signature for Attended Speaker Detection

\textit{Objective:} Conventional EEG-based auditory attention detection (AAD) is achieved by comparing the time-varying speech stimuli and the elicited EEG signals. However, in order to obtain reliable correlation values, these methods necessitate a long decision window, resulting in a long detection latency. Humans have a remarkable ability to recognize and follow a known speaker, regardless of the spoken content. In this paper, we seek to detect the attended speaker among the pre-enrolled speakers from the elicited EEG signals. In this manner, we avoid relying on the speech stimuli for AAD at run-time. In doing so, we propose a novel EEG-based attended speaker detection (E-ASD) task. \textit{Methods:} We encode a speaker's voice with a fixed dimensional vector, known as speaker embedding, and project it to an audio-derived voice signature, which characterizes the speaker's unique voice regardless of the spoken content. We hypothesize that such a voice signature also exists in the listener's brain that can be decoded from the elicited EEG signals, referred to as EEG-derived voice signature. By comparing the audio-derived voice signature and the EEG-derived voice signature, we are able to effectively detect the attended speaker in the listening brain. \textit{Results:} Experiments show that E-ASD can effectively detect the attended speaker from the 0.5s EEG decision windows, achieving 99.78\% AAD accuracy, 99.94\% AUC, and 0.27\% EER. \textit{Conclusion:} We conclude that it is possible to derive the attended speaker's voice signature from the EEG signals so as to detect the attended speaker in a listening brain. \textit{Significance:} We present the first proof of concept for detecting the attended speaker from the elicited EEG signals in a cocktail party environment. The successful implementation of E-ASD marks a non-trivial, but crucial step towards smart hearing aids.

Ripple sparse self-attention for monaural speech enhancement

The use of Transformer represents a recent success in speech enhancement. However, as its core component, self-attention suffers from quadratic complexity, which is computationally prohibited for long speech recordings. Moreover, it allows each time frame to attend to all time frames, neglecting the strong local correlations of speech signals. This study presents a simple yet effective sparse self-attention for speech enhancement, called ripple attention, which simultaneously performs fine- and coarse-grained modeling for local and global dependencies, respectively. Specifically, we employ local band attention to enable each frame to attend to its closest neighbor frames in a window at fine granularity, while employing dilated attention outside the window to model the global dependencies at a coarse granularity. We evaluate the efficacy of our ripple attention for speech enhancement on two commonly used training objectives. Extensive experimental results consistently confirm the superior performance of the ripple attention design over standard full self-attention, blockwise attention, and dual-path attention (Sep-Former) in terms of speech quality and intelligibility.