In recent years, self-supervised learning has excelled for its capacity to learn robust feature representations from unlabelled data. Networks pretrained through self-supervision serve as effective feature extractors for downstream tasks, including Few-Shot Learning. While the evaluation of unsupervised approaches for few-shot learning is well-established in imagery, it is notably absent in acoustics. This study addresses this gap by assessing large-scale self-supervised models' performance in few-shot audio classification. Additionally, we explore the relationship between a model's few-shot learning capability and other downstream task benchmarks. Our findings reveal state-of-the-art performance in some few-shot problems such as SpeechCommandsv2, as well as strong correlations between speech-based few-shot problems and various downstream audio tasks.
Transformers have become central to recent advances in audio classification. However, training an audio spectrogram transformer, e.g. AST, from scratch can be resource and time-intensive. Furthermore, the complexity of transformers heavily depends on the input audio spectrogram size. In this work, we aim to optimize AST training by linking to the resolution in the time-axis. We introduce multi-phase training of audio spectrogram transformers by connecting the seminal idea of coarse-to-fine with transformer models. To achieve this, we propose a set of methods for temporal compression. By employing one of these methods, the transformer model learns from lower-resolution (coarse) data in the initial phases, and then is fine-tuned with high-resolution data later in a curriculum learning strategy. Experimental results demonstrate that the proposed training mechanism for AST leads to improved (or on-par) performance with faster convergence, i.e. requiring fewer computational resources and less time. This approach is also generalizable to other AST-based methods regardless of their learning paradigms.
Conventional audio classification relied on predefined classes, lacking the ability to learn from free-form text. Recent methods unlock learning joint audio-text embeddings from raw audio-text pairs describing audio in natural language. Despite recent advancements, there is little exploration of systematic methods to train models for recognizing sound events and sources in alternative scenarios, such as distinguishing fireworks from gunshots at outdoor events in similar situations. This study introduces causal reasoning and counterfactual analysis in the audio domain. We use counterfactual instances and include them in our model across different aspects. Our model considers acoustic characteristics and sound source information from human-annotated reference texts. To validate the effectiveness of our model, we conducted pre-training utilizing multiple audio captioning datasets. We then evaluate with several common downstream tasks, demonstrating the merits of the proposed method as one of the first works leveraging counterfactual information in audio domain. Specifically, the top-1 accuracy in open-ended language-based audio retrieval task increased by more than 43%.
In this paper, we propose a method for class-incremental learning of potentially overlapping sounds for solving a sequence of multi-label audio classification tasks. We design an incremental learner that learns new classes independently of the old classes. To preserve knowledge about the old classes, we propose a cosine similarity-based distillation loss that minimizes discrepancy in the feature representations of subsequent learners, and use it along with a Kullback-Leibler divergence-based distillation loss that minimizes discrepancy in their respective outputs. Experiments are performed on a dataset with 50 sound classes, with an initial classification task containing 30 base classes and 4 incremental phases of 5 classes each. After each phase, the system is tested for multi-label classification with the entire set of classes learned so far. The proposed method obtains an average F1-score of 40.9% over the five phases, ranging from 45.2% in phase 0 on 30 classes, to 36.3% in phase 4 on 50 classes. Average performance degradation over incremental phases is only 0.7 percentage points from the initial F1-score of 45.2%.
Audio self-supervised learning (SSL) pre-training, which aims to learn good representations from unlabeled audio, has made remarkable progress. However, the extensive computational demands during pre-training pose a significant barrier to the potential application and optimization of audio SSL models. In this paper, inspired by the success of data2vec 2.0 in image modality and Audio-MAE in audio modality, we introduce Efficient Audio Transformer (EAT) to further improve the effectiveness and efficiency in audio SSL. The proposed EAT adopts the bootstrap self-supervised training paradigm to the audio domain. A novel Utterance-Frame Objective (UFO) is designed to enhance the modeling capability of acoustic events. Furthermore, we reveal that the masking strategy is critical in audio SSL pre-training, and superior audio representations can be obtained with large inverse block masks. Experiment results demonstrate that EAT achieves state-of-the-art (SOTA) performance on a range of audio-related tasks, including AudioSet (AS-2M, AS-20K), ESC-50, and SPC-2, along with a significant pre-training speedup up to ~15x compared to existing audio SSL models.
The recognition of underwater audio plays a significant role in identifying a vessel while it is in motion. Underwater target recognition tasks have a wide range of applications in areas such as marine environmental protection, detection of ship radiated noise, underwater noise control, and coastal vessel dispatch. The traditional UATR task involves training a network to extract features from audio data and predict the vessel type. The current UATR dataset exhibits shortcomings in both duration and sample quantity. In this paper, we propose Oceanship, a large-scale and diverse underwater audio dataset. This dataset comprises 15 categories, spans a total duration of 121 hours, and includes comprehensive annotation information such as coordinates, velocity, vessel types, and timestamps. We compiled the dataset by crawling and organizing original communication data from the Ocean Communication Network (ONC) database between 2021 and 2022. While audio retrieval tasks are well-established in general audio classification, they have not been explored in the context of underwater audio recognition. Leveraging the Oceanship dataset, we introduce a baseline model named Oceannet for underwater audio retrieval. This model achieves a recall at 1 (R@1) accuracy of 67.11% and a recall at 5 (R@5) accuracy of 99.13% on the Deepship dataset.
Current state-of-the-art audio analysis systems rely on pre-trained embedding models, often used off-the-shelf as (frozen) feature extractors. Choosing the best one for a set of tasks is the subject of many recent publications. However, one aspect often overlooked in these works is the influence of the duration of audio input considered to extract an embedding, which we refer to as Temporal Support (TS). In this work, we study the influence of the TS for well-established or emerging pre-trained embeddings, chosen to represent different types of architectures and learning paradigms. We conduct this evaluation using both musical instrument and environmental sound datasets, namely OpenMIC, TAU Urban Acoustic Scenes 2020 Mobile, and ESC-50. We especially highlight that Audio Spectrogram Transformer-based systems (PaSST and BEATs) remain effective with smaller TS, which therefore allows for a drastic reduction in memory and computational cost. Moreover, we show that by choosing the optimal TS we reach competitive results across all tasks. In particular, we improve the state-of-the-art results on OpenMIC, using BEATs and PaSST without any fine-tuning.
Despite the remarkable advances in deep learning technology, achieving satisfactory performance in lung sound classification remains a challenge due to the scarcity of available data. Moreover, the respiratory sound samples are collected from a variety of electronic stethoscopes, which could potentially introduce biases into the trained models. When a significant distribution shift occurs within the test dataset or in a practical scenario, it can substantially decrease the performance. To tackle this issue, we introduce cross-domain adaptation techniques, which transfer the knowledge from a source domain to a distinct target domain. In particular, by considering different stethoscope types as individual domains, we propose a novel stethoscope-guided supervised contrastive learning approach. This method can mitigate any domain-related disparities and thus enables the model to distinguish respiratory sounds of the recording variation of the stethoscope. The experimental results on the ICBHI dataset demonstrate that the proposed methods are effective in reducing the domain dependency and achieving the ICBHI Score of 61.71%, which is a significant improvement of 2.16% over the baseline.
Parameter-efficient transfer learning (PETL) methods have emerged as a solid alternative to the standard full fine-tuning approach. They only train a few extra parameters for each downstream task, without sacrificing performance and dispensing with the issue of storing a copy of the pre-trained model for each task. For audio classification tasks, the Audio Spectrogram Transformer (AST) model shows impressive results. However, surprisingly, how to efficiently adapt it to several downstream tasks has not been tackled before. In this paper, we bridge this gap and present a detailed investigation of common PETL methods for the adaptation of the AST model to audio/speech tasks. Furthermore, we propose a new adapter design that exploits the convolution module of the Conformer model, leading to superior performance over the standard PETL approaches and surpassing or achieving performance parity with full fine-tuning by updating only 0.29% of the parameters. Finally, we provide ablation studies revealing that our proposed adapter: 1) proves to be effective in few-shot efficient transfer learning, 2) attains optimal results regardless of the amount of the allocated parameters, and 3) can be applied to other pre-trained models.
The auditory system plays a substantial role in shaping the overall human perceptual experience. While prevailing large language models (LLMs) and visual language models (VLMs) have shown their promise in solving a wide variety of language and vision understanding tasks, only a few of them can be generalised to the audio domain without compromising their domain-specific capability. In this work, we introduce Acoustic Prompt Tuning (APT), a new adapter extending LLMs and VLMs to the audio domain by injecting audio embeddings to the input of LLMs, namely soft prompting. Specifically, APT applies an instruction-aware audio aligner to generate soft prompts, conditioned on both input text and sounds, as the inputs to the language model. To mitigate data scarcity in the audio domain, a curriculum learning strategy is proposed by formulating diverse audio tasks in a sequential manner. Moreover, we improve the audio language model by using interleaved audio-text embeddings as the input sequence. In this improved model, zero constraints are imposed on the input format, thus it is capable of tackling diverse modelling tasks, such as few-shot audio classification and audio comparison. To further evaluate the advanced ability of the audio networks, we introduce natural language audio reasoning (NLAR), a new task that analyses two audio clips by comparison and summarisation. Experiments show that APT-enhanced LLMs (namely APT-LLMs) achieve competitive results compared to the expert models (i.e., the networks trained on the target datasets) across various tasks. We finally demonstrate APT's ability in extending frozen VLMs to the audio domain without fine-tuning, achieving promising results in audio-visual question and answering. Our code and model weights will be released at https://github.com/JinhuaLiang/APT
Formally verifying audio classification systems is essential to ensure accurate signal classification across real-world applications like surveillance, automotive voice commands, and multimedia content management, preventing potential errors with serious consequences. Drawing from recent research, this study advances the utilization of star-set-based formal verification, extended through reachability analysis, tailored explicitly for Long Short-Term Memory architectures and their Convolutional variations within the audio classification domain. By conceptualizing the classification process as a sequence of set operations, the star set-based reachability approach streamlines the exploration of potential operational states attainable by the system. The paper serves as an encompassing case study, validating and verifying sequence audio classification analytics within real-world contexts. It accentuates the necessity for robustness verification to ensure precise and dependable predictions, particularly in light of the impact of noise on the accuracy of output classifications.
Text and vision foundation models can perform many tasks in a zero-shot setting, a desirable property that enables these systems to be applied in general and low-resource settings. There has been far less work, however, on the zero-shot abilities of ASR foundation models, with these systems typically fine-tuned to specific tasks or constrained to applications that match their training criterion and data annotation. In this work we investigate the ability of Whisper and MMS, ASR foundation models trained primarily for speech recognition, to perform zero-shot audio classification. We use simple template-based text prompts at the decoder and use the resulting decoding probabilities to generate zero-shot predictions. Without training the model on extra data or adding any new parameters, we demonstrate that Whisper shows promising zero-shot classification performance on a range of 8 audio-classification datasets, outperforming the accuracy of existing state-of-the-art zero-shot baselines by an average of 9%. One important step to unlock the emergent ability is debiasing, where a simple unsupervised reweighting method of the class probabilities yields consistent significant performance gains. We further show that performance increases with model size, implying that as ASR foundation models scale up, they may exhibit improved zero-shot performance.
Recently, instruction-following audio-language models have received broad attention for audio interaction with humans. However, the absence of pre-trained audio models capable of handling diverse audio types and tasks has hindered progress in this field. Consequently, most existing works have only been able to support a limited range of interaction capabilities. In this paper, we develop the Qwen-Audio model and address this limitation by scaling up audio-language pre-training to cover over 30 tasks and various audio types, such as human speech, natural sounds, music, and songs, to facilitate universal audio understanding abilities. However, directly co-training all tasks and datasets can lead to interference issues, as the textual labels associated with different datasets exhibit considerable variations due to differences in task focus, language, granularity of annotation, and text structure. To overcome the one-to-many interference, we carefully design a multi-task training framework by conditioning on a sequence of hierarchical tags to the decoder for encouraging knowledge sharing and avoiding interference through shared and specified tags respectively. Remarkably, Qwen-Audio achieves impressive performance across diverse benchmark tasks without requiring any task-specific fine-tuning, surpassing its counterparts. Building upon the capabilities of Qwen-Audio, we further develop Qwen-Audio-Chat, which allows for input from various audios and text inputs, enabling multi-turn dialogues and supporting various audio-central scenarios.
Deep generative models have emerged as a promising approach in the medical image domain to address data scarcity. However, their use for sequential data like respiratory sounds is less explored. In this work, we propose a straightforward approach to augment imbalanced respiratory sound data using an audio diffusion model as a conditional neural vocoder. We also demonstrate a simple yet effective adversarial fine-tuning method to align features between the synthetic and real respiratory sound samples to improve respiratory sound classification performance. Our experimental results on the ICBHI dataset demonstrate that the proposed adversarial fine-tuning is effective, while only using the conventional augmentation method shows performance degradation. Moreover, our method outperforms the baseline by 2.24% on the ICBHI Score and improves the accuracy of the minority classes up to 26.58%. For the supplementary material, we provide the code at https://github.com/kaen2891/adversarial_fine-tuning_using_generated_respiratory_sound.
The rapid advancement of artificial intelligence (AI) has been marked by the large language models exhibiting human-like intelligence. However, these models also present unprecedented challenges to energy consumption and environmental sustainability. One promising solution is to revisit analogue computing, a technique that predates digital computing and exploits emerging analogue electronic devices, such as resistive memory, which features in-memory computing, high scalability, and nonvolatility. However, analogue computing still faces the same challenges as before: programming nonidealities and expensive programming due to the underlying devices physics. Here, we report a universal solution, software-hardware co-design using structural plasticity-inspired edge pruning to optimize the topology of a randomly weighted analogue resistive memory neural network. Software-wise, the topology of a randomly weighted neural network is optimized by pruning connections rather than precisely tuning resistive memory weights. Hardware-wise, we reveal the physical origin of the programming stochasticity using transmission electron microscopy, which is leveraged for large-scale and low-cost implementation of an overparameterized random neural network containing high-performance sub-networks. We implemented the co-design on a 40nm 256K resistive memory macro, observing 17.3% and 19.9% accuracy improvements in image and audio classification on FashionMNIST and Spoken digits datasets, as well as 9.8% (2%) improvement in PR (ROC) in image segmentation on DRIVE datasets, respectively. This is accompanied by 82.1%, 51.2%, and 99.8% improvement in energy efficiency thanks to analogue in-memory computing. By embracing the intrinsic stochasticity and in-memory computing, this work may solve the biggest obstacle of analogue computing systems and thus unleash their immense potential for next-generation AI hardware.
Majority of research in learning based methods has been towards designing and training networks for specific tasks. However, many of the learning based tasks, across modalities, share commonalities and could be potentially tackled in a joint framework. We present an approach in such direction, to learn multiple tasks, in multiple modalities, with a unified architecture. The proposed network is composed of task specific encoders, a common trunk in the middle, followed by task specific prediction heads. We first pre-train it by self-supervised masked training, followed by sequential training for the different tasks. We train the network on all major modalities, e.g.\ visual, audio, text and 3D, and report results on $22$ diverse and challenging public benchmarks. We demonstrate empirically that, using a joint network to train across modalities leads to meaningful information sharing and this allows us to achieve state-of-the-art results on most of the benchmarks. We also show generalization of the trained network on cross-modal tasks as well as unseen datasets and tasks.
One of the main challenges of multimodal learning is the need to combine heterogeneous modalities (e.g., video, audio, text). For example, video and audio are obtained at much higher rates than text and are roughly aligned in time. They are often not synchronized with text, which comes as a global context, e.g., a title, or a description. Furthermore, video and audio inputs are of much larger volumes, and grow as the video length increases, which naturally requires more compute dedicated to these modalities and makes modeling of long-range dependencies harder. We here decouple the multimodal modeling, dividing it into separate, focused autoregressive models, processing the inputs according to the characteristics of the modalities. We propose a multimodal model, called Mirasol3B, consisting of an autoregressive component for the time-synchronized modalities (audio and video), and an autoregressive component for the context modalities which are not necessarily aligned in time but are still sequential. To address the long-sequences of the video-audio inputs, we propose to further partition the video and audio sequences in consecutive snippets and autoregressively process their representations. To that end, we propose a Combiner mechanism, which models the audio-video information jointly within a timeframe. The Combiner learns to extract audio and video features from raw spatio-temporal signals, and then learns to fuse these features producing compact but expressive representations per snippet. Our approach achieves the state-of-the-art on well established multimodal benchmarks, outperforming much larger models. It effectively addresses the high computational demand of media inputs by both learning compact representations, controlling the sequence length of the audio-video feature representations, and modeling their dependencies in time.
This study investigates the potential of automated deep learning to enhance the accuracy and efficiency of multi-class classification of bird vocalizations, compared against traditional manually-designed deep learning models. Using the Western Mediterranean Wetland Birds dataset, we investigated the use of AutoKeras, an automated machine learning framework, to automate neural architecture search and hyperparameter tuning. Comparative analysis validates our hypothesis that the AutoKeras-derived model consistently outperforms traditional models like MobileNet, ResNet50 and VGG16. Our approach and findings underscore the transformative potential of automated deep learning for advancing bioacoustics research and models. In fact, the automated techniques eliminate the need for manual feature engineering and model design while improving performance. This study illuminates best practices in sampling, evaluation and reporting to enhance reproducibility in this nascent field. All the code used is available at https: //github.com/giuliotosato/AutoKeras-bioacustic Keywords: AutoKeras; automated deep learning; audio classification; Wetlands Bird dataset; comparative analysis; bioacoustics; validation dataset; multi-class classification; spectrograms.
Artificial neural networks (ANNs) are the basis of recent advances in artificial intelligence (AI); they typically use real valued neuron responses. By contrast, biological neurons are known to operate using spike trains. In principle, spiking neural networks (SNNs) may have a greater representational capability than ANNs, especially for time series such as speech; however their adoption has been held back by both a lack of stable training algorithms and a lack of compatible baselines. We begin with a fairly thorough review of literature around the conjunction of ANNs and SNNs. Focusing on surrogate gradient approaches, we proceed to define a simple but relevant evaluation based on recent speech command tasks. After evaluating a representative selection of architectures, we show that a combination of adaptation, recurrence and surrogate gradients can yield light spiking architectures that are not only able to compete with ANN solutions, but also retain a high degree of compatibility with them in modern deep learning frameworks. We conclude tangibly that SNNs are appropriate for future research in AI, in particular for speech processing applications, and more speculatively that they may also assist in inference about biological function.
The introduction of large-scale audio datasets, such as AudioSet, paved the way for Transformers to conquer the audio domain and replace CNNs as the state-of-the-art neural network architecture for many tasks. Audio Spectrogram Transformers are excellent at exploiting large datasets, creating powerful pre-trained models that surpass CNNs when fine-tuned on downstream tasks. However, current popular Audio Spectrogram Transformers are demanding in terms of computational complexity compared to CNNs. Recently, we have shown that, by employing Transformer-to-CNN Knowledge Distillation, efficient CNNs can catch up with and even outperform Transformers on large datasets. In this work, we extend this line of research and increase the capacity of efficient CNNs by introducing dynamic CNN blocks, constructed of dynamic non-linearities, dynamic convolutions and attention mechanisms. We show that these dynamic CNNs outperform traditional efficient CNNs, in terms of the performance-complexity trade-off and parameter efficiency, at the task of audio tagging on the large-scale AudioSet. Our experiments further indicate that the introduced dynamic CNNs achieve better performance on downstream tasks and scale up well, attaining Transformer performance and even outperforming them on AudioSet and several downstream tasks.