Presents corrections to the article “ARAUS: A Large-Scale Dataset and Baseline Models of Affective Responses to Augmented Urban Soundscapes”.

Spatial audio quality is a highly multifaceted concept, with many interactions between environmental, geometrical, anatomical, psychological, and contextual factors. Methods for characterization or evaluation of the geometrical components of spatial audio quality, however, remain scarce, despite being perhaps the least subjective aspect of spatial audio quality to quantify. By considering interchannel time and level differences relative to a reference signal, it is possible to construct a signal model to isolate some of the spatial distortion. By using a combination of least-square optimization and heuristics, we propose a signal decomposition method to isolate the spatial error, in terms of interchannel gain leakages and changes in relative delays, from a processed signal. This allows the computation of simple energy-ratio metrics, providing objective measures of spatial and non-spatial signal qualities, with minimal assumptions and no dataset dependency. Experiments demonstrate the robustness of the method against common spatial signal degradation introduced by, e.g., audio compression and music source separation.

Choosing optimal maskers for existing soundscapes to effect a desired perceptual change via soundscape augmentation is non-trivial due to extensive varieties of maskers and a dearth of benchmark datasets with which to compare and develop soundscape augmentation models. To address this problem, we make publicly available the ARAUS (Affective Responses to Augmented Urban Soundscapes) dataset, which comprises a five-fold cross-validation set and independent test set totaling 25,440 unique subjective perceptual responses to augmented soundscapes presented as audio-visual stimuli. Each augmented soundscape is made by digitally adding “maskers” (bird, water, wind, traffic, construction, or silence) to urban soundscape recordings at fixed soundscape-to-masker ratios. Responses were then collected by asking participants to rate how pleasant, annoying, eventful, uneventful, vibrant, monotonous, chaotic, calm, and appropriate each augmented soundscape was, in accordance with ISO/TS 12913-2:2018. Participants also provided relevant demographic information and completed standard psychological questionnaires. We perform exploratory and statistical analysis of the responses obtained to verify internal consistency and agreement with known results in the literature. Finally, we demonstrate the benchmarking capability of the dataset by training and comparing four baseline models for urban soundscape pleasantness: a low-parameter regression model, a high-parameter convolutional neural network, and two attention-based networks in the literature.

Autonomous soundscape augmentation systems typically use trained models to pick optimal maskers to effect a desired perceptual change. While acoustic information is paramount to such systems, contextual information, including participant demographics and the visual environment, also influences acoustic perception. Hence, we propose modular modifications to an existing attention-based deep neural network, to allow early, mid-level, and late feature fusion of participant-linked, visual, and acoustic features. Ablation studies on module configurations and corresponding fusion methods using the ARAUS dataset show that contextual features improve the model performance in a statistically significant manner on the normalized ISO Pleasantness, to a mean squared error of 0.1194±0.0012 for the best-performing all-modality model, against 0.1217±0.0009 for the audio-only model. Soundscape augmentation systems can thereby leverage multimodal inputs for improved performance. We also investigate the impact of individual participant-linked factors using trained models to illustrate improvements in model explainability.

Polyphonic sound event localization and detection (SELD) has many practical applications in acoustic sensing and monitoring. However, the development of real-time SELD has been limited by the demanding computational requirement of most recent SELD systems. In this work, we introduce SALSA-Lite, a fast and effective feature for polyphonic SELD using microphone array inputs. SALSA-Lite is a lightweight variation of a previously proposed SALSA feature for polyphonic SELD. SALSA, which stands for Spatial Cue-Augmented Log-Spectrogram, consists of multichannel log-spectrograms stacked channel-wise with the normalized principal eigenvectors of the spectrotemporally corresponding spatial covariance matrices. In contrast to SALSA, which uses eigenvector-based spatial features, SALSA-Lite uses normalized inter-channel phase differences as spatial features, allowing a 30-fold speedup compared to the original SALSA feature. Experimental results on the TAU-NIGENS Spatial Sound Events 2021 dataset showed that the SALSA-Lite feature achieved competitive performance compared to the full SALSA feature, and significantly outperformed the traditional feature set of multichannel log-mel spectrograms with generalized cross-correlation spectra. Specifically, using SALSA-Lite features increased localization-dependent F1 score and class-dependent localization recall by 15% and 5%, respectively, compared to using multichannel log-mel spectrograms with generalized cross-correlation spectra.

Soundscape augmentation, which involves the addition of sounds known as "maskers" to a given soundscape, is a human-centric urban noise mitigation measure aimed at improving the overall sound-scape quality. However, the choice of maskers is often predicated on laborious processes and is inflexible to the time-varying nature of real-world soundscapes. Owing to the perceptual uniqueness of each soundscape and the inherent subjectiveness of human perception, we propose a probabilistic perceptual attribute predictor (PPAP) that predicts parameters of random distributions as outputs instead of a single deterministic value. Using the PPAP, we developed a novel automatic masker selection system (AMSS), which selects optimal masker candidates based on the predicted distribution of the ISO 12913-3 Pleasantness score for a given soundscape. Via a largescale listening test with 300 participants, we collected 12600 subjective responses, each to a unique augmented soundscape, to train the PPAP models in a 5-fold cross-validation scheme. Using a convolutional recurrent neural network backbone and experimenting with several variants of the attention mechanism for the PPAP, we evaluated the proposed system on a blind test set with 48 unseen augmented soundscapes to assess the effectiveness of the probabilistic output scheme over traditional deterministic systems.

Echo and noise suppression is an integral part of a full-duplex communication system. Many recent acoustic echo cancellation (AEC) systems rely on a separate adaptive filtering module for linear echo suppression and a neural module for residual echo suppression. However, in practice, adaptive filtering modules require time to converge and remain susceptible to changes in the acoustic environment. This introduces unnecessary delays to AEC systems using this two-stage framework, despite neural modules already having the capability to suppress both linear and nonlinear echo components. In this paper, we exploit the offset-compensating property of complex time-frequency masks and propose an end-to-end complex-valued neural network architecture. The building block of the proposed model is a pseudocomplex extension of the densely-connected multidilated DenseNet (D3Net), resulting in a very small network of only 354K parameters. The architecture utilized the multi-resolution nature of the D3Net to eliminate the need for pooling, allowing feature extraction using large receptive fields without any loss of output resolution. We also propose a dual-mask technique for joint echo and noise suppression with simultaneous speech enhancement. Evaluation on both synthetic and real test sets demonstrated promising results across multiple energy-based metrics and perceptual proxies.

Convolutional recurrent networks (CRN) integrating a convolutional encoder-decoder (CED) structure and a recurrent structure have achieved promising performance for monaural speech enhancement. However, feature representation across frequency context is highly constrained due to limited receptive fields in the convolutions of CED. In this paper, we propose a convolutional recurrent encoder-decoder (CRED) structure to boost feature representation along the frequency axis. The CRED applies frequency recurrence on 3D convolutional feature maps along the frequency axis following each convolution, therefore, it is capable of catching long-range frequency correlations and enhancing feature representations of speech inputs. The proposed frequency recurrence is realized efficiently using a feedforward sequential memory network (FSMN). Besides the CRED, we insert two stacked FSMN layers between the encoder and the decoder to model further temporal dynamics. We name the proposed framework as Frequency Recurrent CRN (FRCRN). We design FRCRN to predict complex Ideal Ratio Mask (cIRM) in complex-valued domain and optimize FRCRN using both time-frequency-domain and time-domain losses. Our proposed approach achieved state-of-the-art performance on wideband bench-mark datasets and achieved 2nd place for the real-time fullband track in terms of Mean Opinion Score (MOS) and Word Accuracy (WAcc) in the ICASSP 2022 Deep Noise Suppression (DNS) challenge.

The selection of maskers and playback gain levels in an in-situ soundscape augmentation system is crucial to its effectiveness in improving the overall acoustic comfort of a given environment. Traditionally, the selection of appropriate maskers and gain levels has been informed by expert opinion, which may not be representative of the target population, or by listening tests, which can be time- and labor-intensive. Furthermore, the resulting static choices of masker and gain are often inflexible to dynamic real-world soundscapes. In this work, we utilized a deep learning model to perform joint selection of the optimal masker and its gain level for a given soundscape. The proposed model was designed with highly modular building blocks, allowing for an optimized inference process that can quickly search through a large number of masker-gain combinations. In addition, we introduced the use of feature-domain soundscape augmentation conditioned on the digital gain level, eliminating the computationally expensive waveform-domain mixing process during inference, as well as the tedious gain adjustment process required for new maskers. The proposed system was evaluated on a large-scale dataset of subjective responses to augmented soundscapes with 442 participants, with the best model achieving a mean squared error of${0.122}\mathbf {\pm }{0.005}$on pleasantness score, validating the ability of the model to predict combined effect of the masker and its gain level on the perceptual pleasantness level. The proposed system thus allows in-situ or mixed-reality soundscape augmentation to be performed autonomously with near real-time latency while continuously accounting for changes in acoustic environments.

Sound event localization and detection (SELD) consists of two subtasks, which are sound event detection and direction-of-arrival estimation. While sound event detection mainly relies on time-frequency patterns to distinguish different sound classes, direction-of-arrival estimation uses amplitude and/or phase differences between microphones to estimate source directions. As a result, it is often difficult to jointly optimize these two subtasks. We propose a novel feature calledSpatial cue-Augmented Log-SpectrogrAm(SALSA) with exact time-frequency mapping between the signal power and the source directional cues, which is crucial for resolving overlapping sound sources. The SALSA feature consists of multichannel log-spectrograms stacked along with the normalized principal eigenvector of the spatial covariance matrix at each corresponding time-frequency bin. Depending on the microphone array format, the principal eigenvector can be normalized differently to extract amplitude and/or phase differences between the microphones. As a result, SALSA features are applicable for different microphone array formats such as first-order ambisonics (FOA) and multichannel microphone array (MIC). Experimental results on the TAU-NIGENS Spatial Sound Events 2021 dataset with directional interferences showed that SALSA features outperformed other state-of-the-art features. Specifically, the use of SALSA features in the FOA format increased the F1 score and localization recall by$6 \,\%$each, compared to the multichannel log-mel spectrograms with intensity vectors. For the MIC format, using SALSA features increased F1 score and localization recall by$16 \,\%$and$7 \,\%$, respectively, compared to using multichannel log-mel spectrograms with generalized cross-correlation spectra.

In blind source separation of speech signals, the inherent imbalance in the source spectrum poses a challenge for methods that rely on single-source dominance for the estimation of the mixing matrix. We propose an algorithm based on the directional sparse filtering (DSF) framework that utilizes the Lehmer mean with learnable weights to adaptively account for source imbalance. Performance evaluation in multiple real acoustic environments show improvements in source separation compared to the baseline methods.