Description on IEEE ICME 2024 Grand Challenge: Semi-supervised Acoustic Scene Classification under Domain Shift

Acoustic scene classification (ASC) is a crucial research problem in computational auditory scene analysis, and it aims to recognize the unique acoustic characteristics of an environment. One of the challenges of the ASC task is domain shift caused by a distribution gap between training and testing data. Since 2018, ASC challenges have focused on the generalization of ASC models across different recording devices. Although this task in recent years has achieved substantial progress in device generalization, the challenge of domain shift between different regions, involving characteristics such as time, space, culture, and language, remains insufficiently explored at present. In addition, considering the abundance of unlabeled acoustic scene data in the real world, it is important to study the possible ways to utilize these unlabelled data. Therefore, we introduce the task Semi-supervised Acoustic Scene Classification under Domain Shift in the ICME 2024 Grand Challenge. We encourage participants to innovate with semi-supervised learning techniques, aiming to develop more robust ASC models under domain shift.

Sub-band and Full-band Interactive U-Net with DPRNN for Demixing Cross-talk Stereo Music

This paper presents a detailed description of our proposed methods for the ICASSP 2024 Cadenza Challenge. Experimental results show that the proposed system can achieve better performance than official baselines.

Interactive Dual-Conformer with Scene-Inspired Mask for Soft Sound Event Detection

Traditional binary hard labels for sound event detection (SED) lack details about the complexity and variability of sound event distributions. Recently, a novel annotation workflow is proposed to generate fine-grained non-binary soft labels, resulting in a new real-life dataset named MAESTRO Real for SED. In this paper, we first propose an interactive dual-conformer (IDC) module, in which a cross-interaction mechanism is applied to effectively exploit the information from soft labels. In addition, a novel scene-inspired mask (SIM) based on soft labels is incorporated for more precise SED predictions. The SIM is initially generated through a statistical approach, referred as SIM-V1. However, the fixed artificial mask may mismatch the SED model, resulting in limited effectiveness. Therefore, we further propose SIM-V2, which employs a word embedding model for adaptive SIM estimation. Experimental results show that the proposed IDC module can effectively utilize the information from soft labels, and the integration of SIM-V1 can further improve the accuracy. In addition, the impact of different word embedding dimensions on SIM-V2 is explored, and the results show that the appropriate dimension can enable SIM-V2 achieve superior performance than SIM-V1. In DCASE 2023 Challenge Task4B, the proposed system achieved the top ranking performance on the evaluation dataset of MAESTRO Real.

Interactive Dual Attention Conformer With Scene-Based Mask For Soft Sound Event Detection

The emergence of soft-labeled data for sound event detection (SED) effectively overcomes the lack of traditional strong-labeled data. However, the performance of present SED systems based on such soft labels is still unsatisfactory. In this work, we introduce a dual-branch SED model designed to leverage the information within soft labels. Four variations of the interacted convolutional module are presented to investigate the effective mechanism for information interaction. Furthermore, we incorporate the scene-based mask generated by an estimator to directly apply to the prediction of SED models. Experimental results show that the mask estimator can achieve comparable or even better performance than the manually-designed mask and significantly improve the performance of SED. The proposed approach achieved the top ranking in the DCASE 2023 Task4B Challenge.

AudioLog: LLMs-Powered Long Audio Logging with Acoustic Scenes and Events Joint Estimation

Previous studies in automated audio captioning have faced difficulties in accurately capturing the complete temporal details of acoustic scenes and events within long audio sequences. This paper presents AudioLog, a large language models (LLMs)-powered audio logging system with multi-task learning of acoustic tasks. Specifically, we propose a joint training network, achieved by fine-tuning a large audio model based on the pre-trained hierarchical token-semantic audio Transformer. We then leverage LLMs to craft audio logs that summarize textual descriptions of the acoustic environment. Experiments show that the proposed system attains exceptional performance in acoustic scene classification and sound event detection, surpassing existing methods in the field. Further analyses demonstrate AudioLog's power in effectively summarizing long audio sequences.

Two-stage Autoencoder Neural Network for 3D Speech Enhancement

3D speech enhancement has attracted much attention in recent years with the development of augmented reality technology. Traditional denoising convolutional autoencoders have limitations in extracting dynamic voice information. In this paper, we propose a two-stage autoencoder neural network for 3D speech enhancement. We incorporate a dual-path recurrent neural network block into the convolutional autoencoder to iteratively apply time-domain and frequency-domain modeling in an alternate fashion. And an attention mechanism for fusing the high-dimension features is proposed. We also introduce a loss function to simultaneously optimize the network in the time-frequency and time domains. Experimental results show that our system outperforms the state-of-the-art systems on the dataset of ICASSP L3DAS23 challenge.