SEE: Signal Embedding Energy for Quantifying Noise Interference in Large Audio Language Models
Beijing University of Posts and Telecommunications（北京邮电大学）；North China Electric Power University（华北电力大学）；Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；Nanyang Technological University（新加坡南洋理工大学）；Chongqing University of Posts and Telecommunications（重庆邮电大学）
Yuanhe Zhang, Jiayu Tian, Yibo Zhang, Shilinlu Yan, Liang Lin, Zhenhong Zhou, Li Sun, Sen Su
Large Audio Language Models (LALMs) have been widely applied in real-time scenarios, such as in-car assistants and online meeting comprehension. In practice, audio inputs are often corrupted by device and environmental noise, leading to performance degradation. However, existing LALM studies on noise lack quantitative analysis and rely mainly on intuition and empirical observation, thus failing to understand practical robustness. To address this issue, we introduce Signal Embedding Energy (SEE), a method for quantifying the impact of noise intensity on LALM inputs, enabling the differentiation of LALM robustness in real-world deployments. SEE introduces a perspective based on structured activation subspaces derived from the model's internal representations, which more accurately captures its perception of noise than raw audio features. Across experiments, SEE exhibits a strong correlation with LALM performance, achieving a correlation of 0.98. Surprisingly, traditional audio denoising methods are only marginally effective for LALMs, and, in some cases, even increase SEE and impair performance. This suggests a mismatch between speech-centric denoising objectives and the noise sensitivity of modern LALMs. Therefore, we propose a mitigation strategy derived from SEE to denoise LALM inputs, outperforming existing denoising methods. This paper introduces a novel metric for noise quantification in LALMs, providing guidance for robustness improvements in real-world deployments.
In this paper we present a research on identification of audio recording devices from background noise, thus providing a method for forensics. The audio signal is the sum of speech signal and noise signal. Usually, people pay more attention to speech signal, because it carries the information to deliver. So a great amount of researches have been dedicated to getting higher Signal-Noise-Ratio (SNR). There are many speech enhancement algorithms to improve the quality of the speech, which can be seen as reducing the noise. However, noises can be regarded as the intrinsic fingerprint traces of an …
Research in dolphin communication and cognition requires detailed inspection of audible dolphin signals. The manual analysis of these signals is cumbersome and time-consuming. We seek to automate parts of the analysis using modern deep learning methods. We propose to learn an autoencoder constructed from convolutional and recurrent layers trained in an unsupervised fashion. The resulting model embeds patterns in audible dolphin communication. In several experiments, we show that the embeddings can be used for clustering as well as signal detection and signal type classification.
ATRI: Mitigating Multilingual Audio Text Retrieval Inconsistencies by Reducing Data Distribution Errors
Guangdong Provincial Key Laboratory of Ultra High Definition Immersive Media Technology, Shenzhen Graduate School, Peking University（广东 ultra 高清沉浸媒体技术省重点实验室，深圳研究生院，北京大学）；School of Cyber Science and Technology, Sun Yat-sen University（网络安全科学与技术学院，中山大学）；The Chinese University of Hong Kong（香港中文大学）
Isolating the voice of a specific person while filtering out other voices or background noises is challenging when video is shot in noisy environments. We propose audio-visual methods to isolate the voice of a single speaker and eliminate unrelated sounds. First, face motions captured in the video are used to estimate the speaker's voice, by passing the silent video frames through a video-to-speech neural network-based model. Then the speech predictions are applied as a filter on the noisy input audio. This approach avoids using mixtures of sounds in the learning process, as the number of such…
In this paper, we design a system in order to perform the real-time beat tracking for an audio signal. We use Onset Strength Signal (OSS) to detect the onsets and estimate the tempos. Then, we form Cumulative Beat Strength Signal (CBSS) by taking advantage of OSS and estimated tempos. Next, we perform peak detection by extracting the periodic sequence of beats among all CBSS peaks. In simulations, we can see that our proposed algorithm, Online Beat TrAckINg (OBTAIN), outperforms state-of-art results in terms of prediction accuracy while maintaining comparable and practical computational comple…
University of Würzburg（维尔茨堡大学）；Institute of Computer Science, University of Würzburg（维尔茨堡大学计算机科学研究所）
It is easier to hear birds than see them. However, they still play an essential role in nature and are excellent indicators of deteriorating environmental quality and pollution. Recent advances in Deep Neural Networks allow us to process audio data to detect and classify birds. This technology can assist researchers in monitoring bird populations and biodiversity. We propose a sound detection and classification pipeline to analyze complex soundscape recordings and identify birdcalls in the background. Our method learns from weak labels and few data and acoustically recognizes the bird species.…
This paper explores the possibility of using visual object detection techniques for word localization in speech data. Object detection has been thoroughly studied in the contemporary literature for visual data. Noting that an audio can be interpreted as a 1-dimensional image, object localization techniques can be fundamentally useful for word localization. Building upon this idea, we propose a lightweight solution for word detection and localization. We use bounding box regression for word localization, which enables our model to detect the occurrence, offset, and duration of keywords in a giv…
Deep learning-based speech enhancement has seen huge improvements and recently also expanded to full band audio (48 kHz). However, many approaches have a rather high computational complexity and require big temporal buffers for real time usage e.g. due to temporal convolutions or attention. Both make those approaches not feasible on embedded devices. This work further extends DeepFilterNet, which exploits harmonic structure of speech allowing for efficient speech enhancement (SE). Several optimizations in the training procedure, data augmentation, and network structure result in state-of-the-a…
Queen Mary University of London（伦敦玛丽女王大学）；Adobe Research（Adobe研究院）
We present a framework that can impose the audio effects and production style from one recording to another by example with the goal of simplifying the audio production process. We train a deep neural network to analyze an input recording and a style reference recording, and predict the control parameters of audio effects used to render the output. In contrast to past work, we integrate audio effects as differentiable operators in our framework, perform backpropagation through audio effects, and optimize end-to-end using an audio-domain loss. We use a self-supervised training strategy enabling…
Real music signals are highly variable, yet they have strong statistical structure. Prior information about the underlying physical mechanisms by which sounds are generated and rules by which complex sound structure is constructed (notes, chords, a complete musical score), can be naturally unified using Bayesian modelling techniques. Typically algorithms for Automatic Music Transcription independently carry out individual tasks such as multiple-F0 detection and beat tracking. The challenge remains to perform joint estimation of all parameters. We present a Bayesian approach for modelling music…
Thanks to recent advances in deep learning, sophisticated generation tools exist, nowadays, that produce extremely realistic synthetic speech. However, malicious uses of such tools are possible and likely, posing a serious threat to our society. Hence, synthetic voice detection has become a pressing research topic, and a large variety of detection methods have been recently proposed. Unfortunately, they hardly generalize to synthetic audios generated by tools never seen in the training phase, which makes them unfit to face real-world scenarios. In this work, we aim at overcoming this issue by …
Depression detection research has increased over the last few decades, one major bottleneck of which is the limited data availability and representation learning. Recently, self-supervised learning has seen success in pretraining text embeddings and has been applied broadly on related tasks with sparse data, while pretrained audio embeddings based on self-supervised learning are rarely investigated. This paper proposes DEPA, a self-supervised, pretrained depression audio embedding method for depression detection. An encoder-decoder network is used to extract DEPA on in-domain depressed dataset…
In this paper, we seek solutions for reducing the computation complexity of transformer-based models for speech representation learning. We evaluate 10 attention algorithms; then, we pre-train the transformer-based model with those attention algorithms in a self-supervised fashion and treat them as feature extractors on downstream tasks, including phoneme classification and speaker classification. With the assistance of t-SNE, PCA and some observation, the attention weights in self-supervised audio transformers can be categorized into four general cases. Based on these cases and some analyses,…
