Diffusion models have demonstrated remarkable success in generative tasks, including audio super-resolution (SR). In many applications like movie post-production and album mastering, substantial computational budgets are available for achieving superior audio quality. However, while existing diffusion approaches typically increase sampling steps to improve quality, the performance remains fundamentally limited by the stochastic nature of the sampling process, leading to high-variance and quality-limited outputs. Here, rather than simply increasing the number of sampling steps, we propose a different paradigm through inference-time scaling for SR, which explores multiple solution trajectories during the sampling process. Different task-specific verifiers are developed, and two search algorithms, including the random search and zero-order search for SR, are introduced. By actively guiding the exploration of the high-dimensional solution space through verifier-algorithm combinations, we enable more robust and higher-quality outputs. Through extensive validation across diverse audio domains (speech, music, sound effects) and frequency ranges, we demonstrate consistent performance gains, achieving improvements of up to 9.70% in aesthetics, 5.88% in speaker similarity, 15.20% in word error rate, and 46.98% in spectral distance for speech SR from 4 kHz to 24 kHz, showcasing the effectiveness of our approach.

Contrastively pretrained audio–language models (e.g., CLAP) excel at clip-level understanding but struggle with frame-level tasks.Existing extensions fail to exploit the varying granularity of real-world audio–text data, where massive clip-level textual descriptions coexist with limited frame-level annotations. This paper proposes Fine-grained Language-Audio Pretraining (FineLAP), a novel training paradigm that advances both clip- and frame-level alignment in CLAP with heterogeneous data.FineLAP introduces a dual-stream sigmoid loss with a cluster-based sampling strategy to jointly learn from clip- and frame-level supervision. To capture both global semantics and local details, FineLAP uses a decoupled audio projector on top of a self-supervised encoder.To alleviate the scarcity of temporally annotated data, we present FineLAP-100k, a large-scale synthetic SED dataset constructed through a scalable curation pipeline.Extensive experiments demonstrate that FineLAP achieves SOTA performance across multiple audio understanding tasks, including retrieval, classification, sound event detection, and text-to-audio grounding. Ablation studies further show that coarse- and fine-grained alignment are mutually beneficial, providing insights for building better audio-language models (ALMs).

Recent advancements in audio generation have been significantly propelled by the capabilities of Large Language Models (LLMs). The existing research on audio LLM has primarily focused on enhancing the architecture and scale of audio language models, as well as leveraging larger datasets, and generally, acoustic codecs, such as EnCodec, are used for audio tokenization. However, these codecs were originally designed for audio compression, which may lead to suboptimal performance in the context of audio LLM. Our research aims to address the shortcomings of current audio LLM codecs, particularly their challenges in maintaining semantic integrity in generated audio. For instance, existing methods like VALL-E, which condition acoustic token generation on text transcriptions, often suffer from content inaccuracies and elevated word error rates (WER) due to semantic misinterpretations of acoustic tokens, resulting in word skipping and errors. To overcome these issues, we propose a straightforward yet effective approach called X-Codec. X-Codec incorporates semantic features from a pre-trained semantic encoder before the Residual Vector Quantization (RVQ) stage and introduces a semantic reconstruction loss after RVQ. By enhancing the semantic ability of the codec, X-Codec significantly reduces WER in speech synthesis tasks and extends these benefits to non-speech applications, including music and sound generation. Our experiments in text-to-speech, music continuation, and text-to-sound tasks demonstrate that integrating semantic information substantially improves the overall performance of language models in audio generation.

While automated audio captioning (AAC) has made notable progress, traditional fully supervised AAC models still face two critical challenges: the need for expensive audio-text pair data for training and performance degradation when transferring across domains. To overcome these limitations, we present DRCap, a data-efficient and flexible zero-shot audio captioning system that requires text-only data for training and can quickly adapt to new domains without additional fine-tuning. DRCap integrates a contrastive language-audio pre-training (CLAP) model and a large language model (LLM) as its backbone. During training, the model predicts the ground-truth caption with a fixed text encoder from CLAP, whereas, during inference, the text encoder is replaced with the audio encoder to generate captions for audio clips in a zero-shot manner. To mitigate the modality gap of the CLAP model, we use both the projection strategy from the encoder side and the retrieval-augmented generation strategy from the decoder side. Specifically, audio embeddings are first projected onto a text embedding support to absorb extensive semantic information within the joint multi-modal space of CLAP. At the same time, similar captions retrieved from a datastore are fed as prompts to instruct the LLM, incorporating external knowledge to take full advantage of its strong generative capability. Conditioned on both the projected CLAP embedding and the retrieved similar captions, the model is able to produce a more accurate and semantically rich textual description. By tailoring the text embedding support and the caption datastore to the target domain, DRCap acquires a robust ability to adapt to new domains in a training-free manner. Experimental results demonstrate that DRCap outperforms all other zero-shot models in in-domain scenarios and achieves state-of-the-art performance in cross-domain scenarios.

Automatic Music Transcription (AMT), aiming to get musical notes from raw audio, typically uses frame-level systems with piano-roll outputs or language model (LM)-based systems with note-level predictions. However, frame-level systems require manual thresholding, while the LM-based systems struggle with long sequences. In this paper, we propose a hybrid method combining pre-trained roll-based encoders with an LM decoder to leverage the strengths of both methods. Besides, our approach employs a hierarchical prediction strategy, first predicting onset and pitch, then velocity, and finally offset. The hierarchical prediction strategy reduces computational costs by breaking down long sequences into different hierarchies. Evaluated on two benchmark roll-based encoders, our method outperforms traditional piano-roll outputs 0.01 and 0.022 in onset-offset-velocity F1 score, demonstrating its potential as a performance-enhancing plug-in for arbitrary roll-based music transcription encoder. We release the code of this work at https://github.com/yongyizang/AMT_train

Language-queried target sound extraction (TSE) aims to extract specific sounds from mixtures based on language queries. Traditional fully-supervised training schemes require extensively annotated parallel audio-text data, which are labor-intensive. We introduce a parallel-data-free training scheme, requiring only unlabelled audio clips for TSE model training by utilizing the contrastive language-audio pre-trained model (CLAP). In a vanilla parallel-data-free training stage, target audio is encoded using the pre-trained CLAP audio encoder to form a condition embedding, while during testing, user language queries are encoded by CLAP text encoder as the condition embedding. This vanilla approach assumes perfect alignment between text and audio embeddings, which is unrealistic. Two major challenges arise from training-testing mismatch: the persistent modality gap between text and audio and the risk of overfitting due to the exposure of rich acoustic details in target audio embedding during training. To address this, we propose a retrieval-augmented strategy. Specifically, we create an embedding cache using audio captions generated by a large language model (LLM). During training, target audio embeddings retrieve text embeddings from this cache to use as condition embeddings, ensuring consistent modalities between training and testing and eliminating information leakage. Extensive experiment results show that our retrieval-augmented approach achieves consistent and notable performance improvements over existing state-of-the-art with better generalizability.

Non-Intrusive Load Monitoring (NILM), also known as energy disaggregation, aims to infer appliance-level energy consumption from aggregated readings obtained from a single meter measuring a household’s total electricity demand. The increasing availability of large-scale and fine-grained household electricity usage datasets has facilitated the adoption of deep learning techniques for NILM, leading to state-of-the-art performance. However, a key limitation of current approaches is the requirement to train separate models for each appliance, which impedes both model scalability and the ability to integrate new appliances. To address this limitation, we propose a query-based NILM model that leverages appliance embeddings as queries, enabling joint training for multi-appliance disaggregation. By encoding appliance energy usage patterns as embeddings, the proposed model can be queried with these embeddings to extract the energy consumption of corresponding appliances from the aggregated readings and facilitate zero-shot disaggregation. Experimental results on the UK-DALE dataset demonstrate the efficacy of the proposed query-based NILM model.

We introduce Music Source Restoration (MSR), a novel task addressing the gap between idealized source separation and real-world music production. Current Music Source Separation (MSS) approaches assume mixtures are simple sums of sources, ignoring signal degradations employed during music production like equalization, compression, and reverb. MSR models mixtures as degraded sums of individually degraded sources, with the goal of recovering original, undegraded signals. Due to the lack of data for MSR, we present RawStems, a dataset annotation of 578 songs with unprocessed source signals organized into 8 primary and 17 secondary instrument groups, totaling 354.13 hours. To the best of our knowledge, RawStems is the first dataset that contains unprocessed music stems with hierarchical categories. We consider spectral filtering, dynamic range compression, harmonic distortion, reverb and lossy codec as possible degradations, and establish U-Former as a baseline method, demonstrating the feasibility of MSR on our dataset. We release the RawStems dataset annotations, degradation simulation pipeline, training code and pre-trained models to be publicly available.

The audio spectrogram is a time-frequency representation that has been widely used for audio classification. One of the key attributes of the audio spectrogram is the temporal resolution, which depends on the hop size used in the Short-Time Fourier Transform (STFT). Previous works generally assume the hop size should be a constant value (e.g., 10 ms). However, a fixed temporal resolution is not always optimal for different types of sound. The temporal resolution affects not only classification accuracy but also computational cost. This paper proposes a novel method, DiffRes, that enables differentiable temporal resolution modeling for audio classification. Given a spectrogram calculated with a fixed hop size, DiffRes merges non-essential time frames while preserving important frames. DiffRes acts as a "drop-in" module between an audio spectrogram and a classifier and can be jointly optimized with the classification task. We evaluate DiffRes on five audio classification tasks, using mel-spectrograms as the acoustic features, followed by off-the-shelf classifier backbones. Compared with previous methods using the fixed temporal resolution, the DiffRes-based method can achieve the equivalent or better classification accuracy with at least 25% computational cost reduction. We further show that DiffRes can improve classification accuracy by increasing the temporal resolution of input acoustic features, without adding to the computational cost.

Music tagging is a task to predict the tags of music recordings. However, previous music tagging research primarily focuses on close-set music tagging tasks which can not be generalized to new tags. In this work, we propose a zero-shot music tagging system modeled by a joint music and language attention (JMLA) model to address the open-set music tagging problem. The JMLA model consists of an audio encoder modeled by a pretrained masked autoencoder and a decoder modeled by a Falcon7B. We introduce preceiver resampler to convert arbitrary length audio into fixed length embeddings. We introduce dense attention connections between encoder and decoder layers to improve the information flow between the encoder and decoder layers. We collect a large-scale music and description dataset from the internet. We propose to use ChatGPT to convert the raw descriptions into formalized and diverse descriptions to train the JMLA models. Our proposed JMLA system achieves a zero-shot audio tagging accuracy of 64.82% on the GTZAN dataset, outperforming previous zero-shot systems and achieves comparable results to previous systems on the FMA and the MagnaTagATune datasets.

Instrument playing techniques (IPTs) constitute a pivotal component of musical expression. However, the development of automatic IPT detection methods suffers from limited labeled data and inherent class imbalance issues. In this paper, we propose to apply a self-supervised learning model pre-trained on large-scale unlabeled music data and finetune it on IPT detection tasks. This approach addresses data scarcity and class imbalance challenges. Recognizing the significance of pitch in capturing the nuances of IPTs and the importance of onset in locating IPT events, we investigate multi-task finetuning with pitch and onset detection as auxiliary tasks. Additionally, we apply a post-processing approach for event-level prediction, where an IPT activation initiates an event only if the onset output confirms an onset in that frame. Our method outperforms prior approaches in both frame-level and event-level metrics across multiple IPT benchmark datasets. Further experiments demonstrate the efficacy of multi-task finetuning on each IPT class.1

Music source separation (MSS) aims to separate a music recording into multiple musically distinct stems, such as vocals, bass, drums, and more. Recently, deep learning approaches such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) have been used, but the improvement is still limited. In this paper, we propose a novel frequency-domain approach (called BS-RoFormer) based on a Band-Split RoPE Transformer architecture. BS-RoFormer relies on a band-split module to project the input complex spectrogram into subband-level representations, and then arranges a stack of hierarchical Transformers to model the inner-band as well as inter-band sequences for multi-band mask estimation. To facilitate training the model for MSS, we propose to use the Rotary Position Embedding (RoPE). The BS-RoFormer system trained on MUSDB18HQ and 500 extra songs ranked the first place in the Music Separation contest of Sound Demixing Challenge (SDX’23). Benchmarking a smaller version of BS-RoFormer on MUSDB18HQ, we achieve state-of-the-art result without extra training data, with 9.80 dB of average SDR.

The front-end is a critical component of English text-to-speech (TTS) systems, responsible for extracting linguistic features that are essential for a text-to-speech model to synthesize speech, such as prosodies and phonemes. The English TTS front-end typically consists of a text normalization (TN) module, a prosody word prosody phrase (PWPP) module, and a grapheme-to-phoneme (G2P) module. However, current research on the English TTS front-end focuses solely on individual modules, neglecting the interdependence between them and resulting in sub-optimal performance for each module. Therefore, this paper proposes a unified frontend framework that captures the dependencies among the English TTS front-end modules. Extensive experiments have demonstrated that the proposed method achieves state-of-the-art (SOTA) performance in all modules.

Environment shifts and conflicts present significant challenges for learning-based sound event localization and detection (SELD) methods. SELD systems, when trained in particular acoustic settings, often show restricted generalization capabilities for diverse acoustic environments. Furthermore, obtaining annotated samples for spatial sound events is notably costly. Deploying a SELD system in a new environment requires extensive time for re-training and fine-tuning. To overcome these challenges, we propose environment-adaptive Meta-SELD, designed for efficient adaptation to new environments using minimal data. Our method specifically utilizes computationally synthesized spatial data and employs Model-Agnostic Meta-Learning (MAML) on a pre-trained, environment-independent model. The method then utilizes fast adaptation to unseen real-world environments using limited samples from the respective environments. Inspired by the Learning-to-Forget approach, we introduce the concept of selective memory as a strategy for resolving conflicts across environments. This approach involves selectively memorizing target-environment-relevant information and adapting to the new environments through the selective attenuation of model parameters. In addition, we introduce environment representations to characterize different acoustic settings, enhancing the adaptability of our attenuation approach to various environments. We evaluate our proposed method on the development set of the Sony-TAu Realistic Spatial Soundscapes 2023 (STARSS23) dataset and computationally synthesized scenes. Experimental results demonstrate the superior performance of the proposed method compared to conventional supervised learning methods, particularly in localization.

Although audio generation shares commonalities across different types of audio, such as speech, music, and sound effects, designing models for each type requires careful consideration of specific objectives and biases that can significantly differ from those of other types. To bring us closer to a unified perspective of audio generation, this paper proposes a holistic framework that utilizes the same learning method for speech, music, and sound effect generation. Our framework utilizes a general representation of audio, called “language of audio” (LOA). Any audio can be translated into LOA based on AudioMAE, a self-supervised pre-trained representation learning model. In the generation process, we translate other modalities into LOA by using a GPT-2 model, and we perform self-supervised audio generation learning with a latent diffusion model conditioned on the LOA of audio in our training set. The proposed framework naturally brings advantages such as reusable self-supervised pretrained latent diffusion models. Experiments on the major benchmarks of text-to-audio, text-to-music, and text-to-speech with three AudioLDM 2 variants demonstrate competitive performance of the AudioLDM 2 variants framework against previous approaches. Our code, pretrained model, and demo are available athttps://audioldm.github.io/audioldm2.

The advancement of audio-language (AL) multimodal learning tasks has been significant in recent years, yet the limited size of existing audio-language datasets poses challenges for researchers due to the costly and time-consuming collection process. To address this data scarcity issue, we introduceWavCaps, the first large-scale weakly-labelled audio captioning dataset, comprising approximately 400 k audio clips with paired captions. We sourced audio clips and their raw descriptions from web sources and a sound event detection dataset. However, the online-harvested raw descriptions are highly noisy and unsuitable for direct use in tasks such as automated audio captioning. To overcome this issue, we propose a three-stage processing pipeline for filtering noisy data and generating high-quality captions, where ChatGPT, a large language model, is leveraged to filter and transform raw descriptions automatically. We conduct a comprehensive analysis of the characteristics of WavCaps dataset and evaluate it on multiple downstream audio-language multimodal learning tasks. The systems trained on WavCaps outperform previous state-of-the-art (SOTA) models by a significant margin. Our aspiration is for the WavCaps dataset we have proposed to facilitate research in audio-language multimodal learning and demonstrate the potential of utilizing large language models (LLMs) to enhance academic research.

Zero-shot voice conversion (VC) converts source speech into the voice of any desired speaker using only one utterance of the speaker without requiring additional model updates. Typical methods use a speaker representation from a pre-trained speaker verification (SV) model or learn speaker representation during VC training to achieve zero-shot VC. However, existing speaker modeling methods overlook the variation of speaker information richness in temporal and frequency channel dimensions of speech. This insufficient speaker modeling hampers the ability of the VC model to accurately represent unseen speakers who are not in the training dataset. In this study, we present a robust zero-shot VC model withmulti-leveltemporal-channelretrieval, referred to as MTCR-VC. Specifically, to flexibly adapt to the dynamic-variant speaker characteristic in the temporal and channel axis of the speech, we propose a novel fine-grained speaker modeling method, calledtemporal-channelretrieval (TCR), to find outwhenandwherespeaker information appears in speech. It retrieves variable-length speaker representation from both temporal and channel dimensions under the guidance of a pre-trained SV model. Besides, inspired by the hierarchical process of human speech production, the MTCR speaker module stacks several TCR blocks to extract speaker representations from multi-granularity levels. Furthermore, we introduce a cycle-based training strategy to simulate zero-shot inference recurrently to achieve better speech disentanglement and reconstruction. To drive this process, we adopt perceptual constraints on three aspects: content, style, and speaker. Experiments demonstrate that MTCR-VC is superior to the previous zero-shot VC methods in modeling speaker timbre while maintaining good speech naturalness.

Streaming voice conversion (VC) is the task of converting the voice of one person to another in real-time. Previous streaming VC methods use phonetic posteriorgrams (PPGs) extracted from automatic speech recognition (ASR) systems to represent speaker-independent information. However, PPGs lack the prosody and phonation information of the source speaker, and streaming PPGs contain undesired leaked timbre of the source speaker. In this paper, we propose to use intermediate bottleneck features (IBFs) to replace PPGs. VC systems trained with IBFs retain more prosody and phonation information of the source speaker. Furthermore, we propose a non-streaming teacher guidance (TG) framework that addresses the timbre leakage problem. Experiments show that our proposed IBFs and the TG framework achieve a state-of-the-art streaming VC naturalness of 3.85, a content consistency of 3.77, and a timbre similarity of 3.77 under a future receptive field of 160 ms which significantly outperform previous streaming VC systems.

Recently, there has been increasing interest in building efficient audio neural networks for on-device scenarios. Most existing approaches are designed to reduce the size of audio neural networks using methods such as model pruning. In this work, we show that instead of reducing model size using complex methods, eliminating the temporal redundancy in the input audio features (e.g., mel-spectrogram) could be an effective approach for efficient audio classification. To do so, we proposed a family of simple pooling front-ends (SimPFs) which use simple non-parametric pooling operations to reduce the redundant information within the mel-spectrogram. We perform extensive experiments on four audio classification tasks to evaluate the performance of SimPFs. Experimental results show that SimPFs can achieve a reduction in more than half of the number of floating point operations (FLOPs) for off-the-shelf audio neural networks, with negligible degradation or even some improvements in audio classification performance.

Audio captioning aims to generate text descriptions of audio clips.In the real world, many objects produce similar sounds.How to accurately recognize ambiguous sounds is a major challenge for audio captioning.In this work, inspired by inherent human multimodal perception, we propose visuallyaware audio captioning, which makes use of visual information to help the description of ambiguous sounding objects.Specifically, we introduce an off-the-shelf visual encoder to extract video features and incorporate the visual features into an audio captioning system.Furthermore, to better exploit complementary audio-visual contexts, we propose an audio-visual attention mechanism that adaptively integrates audio and visual context and removes the redundant information in the latent space.Experimental results on AudioCaps, the largest audio captioning dataset, show that our proposed method achieves state-of-theart results on machine translation metrics.

This study defines a new evaluation metric for audio tagging tasks to alleviate the limitation of the mean average precision (mAP) metric.The mAP metric treats different kinds of sound as independent classes without considering their relations.The proposed metric, ontology-aware mean average precision (OmAP), addresses the weaknesses of mAP by utilizing additional ontology during evaluation.Specifically, we reweight the false positive events in the model prediction based on the AudioSet ontology graph distance to the target classes.The OmAP also provides insights into model performance by evaluating different coarse-grained levels in the ontology graph.We conduct a human assessment and show that OmAP is more consistent with human perception than mAP.We also propose an ontology-based loss function (OBCE) that reweights binary cross entropy (BCE) loss based on the ontology distance.Our experiment shows that OBCE can improve both mAP and OmAP metrics on the AudioSet tagging task.

Polyphonic sound event localization and detection (SELD) aims at detecting types of sound events with corresponding temporal activities and spatial locations. In this paper, a trackwise ensemble event independent network with a novel data augmentation method is proposed. The proposed model is based on our previous proposed Event-Independent Network V2 and is extended by conformer blocks and dense blocks. The track-wise ensemble model with track-wise output format is proposed to solve an ensemble model problem for track-wise output format that track permutation may occur among different models. The data augmentation approach contains several data augmentation chains, which are composed of random combinations of several data augmentation operations. The method also utilizes log-mel spectrograms, intensity vectors, and Spatial Cues-Augmented Log-Spectrogram (SALSA) for different models. We evaluate our proposed method in the Task of the L3DAS22 challenge and obtain the top ranking solution with a location-dependent F-score to be 0.699. Source code is released1.

Speech super-resolution (SR) is a task to increase speech sampling rate by generating high-frequency components. Existing speech SR methods are trained in constrained experimental settings, such as a fixed upsampling ratio. These strong constraints can potentially lead to poor generalization ability in mismatched real-world cases. In this paper, we propose a neural vocoder based speech super-resolution method (NVSR) that can handle a variety of input resolution and upsampling ratios. NVSR consists of a mel-bandwidth extension module, a neural vocoder module, and a post-processing module. Our proposed system achieves state-of-the-art results on the VCTK multi-speaker benchmark. On 44.1 kHz target resolution, NVSR outperforms WSRGlow and Nu-wave by 8% and 37% respectively on log spectral distance and achieves a significantly better perceptual quality. We also demonstrate that prior knowledge in the pre-trained vocoder is crucial for speech SR by performing mel-bandwidth extension with a simple replication-padding method. Samples can be found in https://haoheliu.github.io/nvsr.

In this paper, we introduce the task of language-queried audio source separation (LASS), which aims to separate a target source from an audio mixture based on a natural language query of the target source (e.g., "a man tells a joke followed by people laughing"). A unique challenge in LASS is associated with the complexity of natural language description and its relation with the audio sources. To address this issue, we proposed LASS-Net, an end-to-end neural network that is learned to jointly process acoustic and linguistic information, and separate the target source that is consistent with the language query from an audio mixture. We evaluate the performance of our proposed system with a dataset created from the AudioCaps dataset. Experimental results show that LASS-Net achieves considerable improvements over baseline methods. Furthermore, we observe that LASS-Net achieves promising generalization results when using diverse human-annotated descriptions as queries, indicating its potential use in real-world scenarios. The separated audio samples and source code are available at https://liuxubo717.github.io/LASS-demopage.

Speech restoration aims to remove distortions in speech signals. Prior methods mainly focus on a single type of distortion, such as speech denoising or dereverberation. However, speech signals can be degraded by several different distortions simultaneously in the real world. It is thus important to extend speech restoration models to deal with multiple distortions. In this paper, we introduce VoiceFixer, a unified framework for high-fidelity speech restoration. VoiceFixer restores speech from multiple distortions (e.g., noise, reverberation, and clipping) and can expand degraded speech (e.g., noisy speech) with a low bandwidth to 44.1 kHz full-bandwidth high-fidelity speech. We design VoiceFixer based on (1) an analysis stage that predicts intermediate-level features from the degraded speech, and (2) a synthesis stage that generates waveform using a neural vocoder. Both objective and subjective evaluations show that VoiceFixer is effective on severely degraded speech, such as real-world historical speech recordings. Samples of VoiceFixer are available at https://haoheliu.github.io/voicefixer.

This letter presents a feature alignment method for domain adaptive Acoustic Scene Classification (ASC) across recording devices. First, we design a two-stream network, in which each stream processes two features,i.e., Log-Mel spectrogram and delta-deltas, using two sub-networks. Second, we investigate different loss functions for feature alignment between the feature maps obtained by the source and target domains. Last, we present an alternate training strategy to deal with the data imbalance problem between paired and unpaired samples. The experimental results obtained on the DCASE benchmarks demonstrate the effectiveness and superiority of the proposed method. The source code of the proposed method is available athttps://github.com/Jingqiao-Zhao/FAASC.

Polyphonic sound event localization and detection (SELD), which jointly performs sound event detection (SED) and direction-of-arrival (DoA) estimation, detects the type and occurrence time of sound events as well as their corresponding DoA angles simultaneously. We study the SELD task from a multi-task learning perspective. Two open problems are addressed in this paper. Firstly, to detect overlapping sound events of the same type but with different DoAs, we propose to use a trackwise output format and solve the accompanying track permutation problem with permutation-invariant training. Multi-head self-attention is further used to separate tracks. Secondly, a previous finding is that, by using hard parameter-sharing, SELD suffers from a performance loss compared with learning the subtasks separately. This is solved by a soft parameter-sharing scheme. We term the proposed method as Event Independent Network V2 (EINV2), which is an improved version of our previously-proposed method and an end-to-end network for SELD. We show that our proposed EINV2 for joint SED and DoA estimation outperforms previous methods by a large margin, and has comparable performance to state-of-the-art ensemble models.

Convolutional neural network (CNN) based methods have achieved state-of-the-art performance for singing melody extraction from polyphonic music. However, most of these methods focus on the learning of local features, while relationships among spectral components locating far apart are often neglected. In this paper, we explore the idea of modeling spectral correlation explicitly for melody extraction. Specifically, we present a spectral correlation module (SCM) that can learn to model the relationships among all frequency bands in a time-frequency representation, thus allowing the encoding of global spectral information into a conventional CNN. Furthermore, we propose to integrate center frequencies with the input feature map of SCM to improve the performance. We implement a light-weight model comprised of SCM blocks to verify the efficacy of our system. Our system achieves a state-of-the-art overall accuracy of 83.5% on the MedleyDB dataset.

Speech enhancement is a task to improve the intelligibility and perceptual quality of degraded speech signal.Recently, neural networks based methods have been applied to speech enhancement.However, many neural network based methods require noisy and clean speech pairs for training.We propose a speech enhancement framework that can be trained with large-scale weakly labelled AudioSet dataset.Weakly labelled data only contain audio tags of audio clips, but not the onset or offset times of speech.We first apply pretrained audio neural networks (PANNs) to detect anchor segments that contain speech or sound events in audio clips.Then, we randomly mix two detected anchor segments containing speech and sound events as a mixture, and build a conditional source separation network using PANNs predictions as soft conditions for speech enhancement.In inference, we input a noisy speech signal with the one-hot encoding of "Speech" as a condition to the trained system to predict enhanced speech.Our system achieves a PESQ of 2.28 and an SSNR of 8.75 dB on the VoiceBank-DEMAND dataset, outperforming the previous SEGAN system of 2.16 and 7.73 dB respectively.

Automatic music transcription (AMT) is the task of transcribing audio recordings into symbolic representations. Recently, neural network based methods have been applied to AMT, and have achieved state-of-the-art results. However, many previous systems only detect onset and offset of notes in frame-wise, so the transcription resolution is limited to the frame hop size. There is a lack of research of using different strategies to encode onset and offset targets for training. In addition, previous AMT systems are sensitive to the misaligned onset and offset labels of audio recordings. Furthermore, there are limited research of sustain pedal transcription on large-scale datasets. In this article, we propose a high-resolution AMT system trained by regressing precise onset and offset times of piano notes. At inference, we propose an algorithm to analytically calculate the precise onset and offset times of piano notes and pedal events. We show that our AMT system is robust to misaligned onset and offset labels compared to previous systems. Our proposed system achieves an onset F1 of 96.72% on the MAESTRO dataset, outperforming previous onsets and frames system of 94.80%. Our system achieves a pedal onset F1 score of 91.86%, which is the first benchmark result on the MAESTRO dataset. We have released the source code and checkpoints of our work at https://github.com/bytedance/piano_transcription.

Energy disaggregation, a.k.a. Non-Intrusive Load Monitoring, aims to separate the energy consumption of individual appliances from the readings of a mains power meter measuring the total energy consumption of, e.g., a whole house. Energy consumption of individual appliances can be useful in many applications, e.g., providing appliance-level feedback to the end users to help them understand their energy consumption and ultimately save energy. Recently, with the availability of large-scale energy consumption datasets, various neural network models such as convolutional neural networks and recurrent neural networks have been investigated to solve the energy disaggregation problem. Neural network models can learn complex patterns from large amounts of data and have been shown to outperform the traditional machine learning methods such as variants of hidden Markov models. However, current neural network methods for energy disaggregation are either computational expensive or are not capable of handling long-term dependencies. In this article, we investigate the application of the recently developed WaveNet models for the task of energy disaggregation. Based on a real-world energy dataset collected from 20 households over 2 years, we show that WaveNet models outperforms the state-of-the-art deep learning methods proposed in the literature for energy disaggregation in terms of both error measures and computational cost. On the basis of energy disaggregation, we then investigate the performance of two deep-learning based frameworks for the task of on/off detection which aims at estimating whether an appliance is in operation or not. The first framework obtains the on/off states of an appliance by binarising the predictions of a regression model trained for energy disaggregation, while the second framework obtains the on/off states of an appliance by directly training a binary classifier with binarised energy readings of the appliance serving as the target values. Based on the same dataset, we show that for the task of on/off detection the second framework, i.e., directly training a binary classifier, achieves better performance in terms of F1 score.

Acoustic Scene Classification (ASC) aims to classify the environment in which the audio signals are recorded. Recently, Convolutional Neural Networks (CNNs) have been successfully applied to ASC. However, the data distributions of the audio signals recorded with multiple devices are different. There has been little research on the training of robust neural networks on acoustic scene datasets recorded with multiple devices, and on explaining the operation of the internal layers of the neural networks. In this article, we focus on training and explaining device-robust CNNs on multi-device acoustic scene data. We propose conditional atrous CNNs with attention for multi-device ASC. Our proposed system contains an ASC branch and a device classification branch, both modelled by CNNs. We visualise and analyse the intermediate layers of the atrous CNNs. A time-frequency attention mechanism is employed to analyse the contribution of each time-frequency bin of the feature maps in the CNNs. On the Detection and Classification of Acoustic Scenes and Events (DCASE) 2018 ASC dataset, recorded with three devices, our proposed model performs significantly better than CNNs trained on single-device data.

In supervised machine learning, the assumption that training data is labelled correctly is not always satisfied. In this paper, we investigate an instance of labelling error for classification tasks in which the dataset is corrupted with out-of-distribution (OOD) instances: data that does not belong to any of the target classes, but is labelled as such. We show that detecting and relabelling certain OOD instances, rather than discarding them, can have a positive effect on learning. The proposed method uses an auxiliary classifier, trained on data that is known to be in-distribution, for detection and relabelling. The amount of data required for this is shown to be small. Experiments are carried out on the FSDnoisy18k audio dataset, where OOD instances are very prevalent. The proposed method is shown to improve the performance of convolutional neural networks by a significant margin. Comparisons with other noise-robust techniques are similarly encouraging.

Source separation is the task of separating an audio recording into individual sound sources. Source separation is fundamental for computational auditory scene analysis. Previous work on source separation has focused on separating particular sound classes such as speech and music. Much previous work requires mixtures and clean source pairs for training. In this work, we propose a source separation framework trained with weakly labelled data. Weakly labelled data only contains the tags of an audio clip, without the occurrence time of sound events. We first train a sound event detection system with AudioSet. The trained sound event detection system is used to detect segments that are most likely to contain a target sound event. Then a regression is learnt from a mixture of two randomly selected segments to a target segment conditioned on the audio tagging prediction of the target segment. Our proposed system can separate 527 kinds of sound classes from AudioSet within a single system. A U-Net is adopted for the separation system and achieves an average SDR of 5.67 dB over 527 sound classes in AudioSet.

Audio pattern recognition is an important research topic in the machine learning area, and includes several tasks such as audio tagging, acoustic scene classification, music classification, speech emotion classification and sound event detection. Recently, neural networks have been applied to tackle audio pattern recognition problems. However, previous systems are built on specific datasets with limited durations. Recently, in computer vision and natural language processing, systems pretrained on large-scale datasets have generalized well to several tasks. However, there is limited research on pretraining systems on large-scale datasets for audio pattern recognition. In this paper, we propose pretrained audio neural networks (PANNs) trained on the large-scale AudioSet dataset. These PANNs are transferred to other audio related tasks. We investigate the performance and computational complexity of PANNs modeled by a variety of convolutional neural networks. We propose an architecture called Wavegram-Logmel-CNN using both log-mel spectrogram and waveform as input feature. Our best PANN system achieves a state-of-the-art mean average precision (mAP) of 0.439 on AudioSet tagging, outperforming the best previous system of 0.392. We transfer PANNs to six audio pattern recognition tasks, and demonstrate state-of-the-art performance in several of those tasks. We have released the source code and pretrained models of PANNs:https://github.com/qiuqiangkong/audioset_tagging_cnn.

Sound event detection (SED) is a task to detect sound events in an audio recording. One challenge of the SED task is that many datasets such as the Detection and Classification of Acoustic Scenes and Events (DCASE) datasets are weakly labelled. That is, there are only audio tags for each audio clip without the onset and offset times of sound events. We compare segment-wise and clip-wise training for SED that is lacking in previous works. We propose a convolutional neural network transformer (CNN-Transfomer) for audio tagging and SED, and show that CNN-Transformer performs similarly to a convolutional recurrent neural network (CRNN). Another challenge of SED is that thresholds are required for detecting sound events. Previous works set thresholds empirically, and are not an optimal approaches. To solve this problem, we propose an automatic threshold optimization method. The first stage is to optimize the system with respect to metrics that do not depend on thresholds, such as mean average precision (mAP). The second stage is to optimize the thresholds with respect to metrics that depends on those thresholds. Our proposed automatic threshold optimization system achieves a state-of-the-art audio tagging F1 of 0.646, outperforming that without threshold optimization of 0.629, and a sound event detection F1 of 0.584, outperforming that without threshold optimization of 0.564.

High quality labeled datasets have allowed deep learning to achieve impressive results on many sound analysis tasks. Yet, it is labor-intensive to accurately annotate large amount of audio data, and the dataset may contain noisy labels in the practical settings. Meanwhile, the deep neural networks are susceptive to those incorrect labeled data because of their outstanding memorization ability. In this article, we present a novel framework, named CrossFilter, to combat the noisy labels problem for audio tagging. Multiple representations (such as, Logmel and MFCC) are used as the input of our framework for providing more complementary information of the audio. Then, though the cooperation and interaction of two neural networks, we divide the dataset into curated and noisy subsets by incrementally pick out the possibly correctly labeled data from the noisy data. Moreover, our approach leverages the multi-task learning on curated and noisy subsets with different loss function to fully utilize the entire dataset. The noisy-robust loss function is employed to alleviate the adverse effects of incorrect labels. On both the audio tagging datasets FSDKaggle2018 and FSDKaggle2019, empirical results demonstrate the performance improvement compared with other competing approaches. On FSDKaggle2018 dataset, our method achieves state-of-the-art performance and even surpasses the ensemble models.

Sound event detection (SED) methods typically rely on either strongly labelled data or weakly labelled data. As an alternative, sequentially labelled data (SLD) was proposed. In SLD, the events and the order of events in audio clips are known, without knowing the occurrence time of events. This paper proposes a connectionist temporal classification (CTC) based SED system that uses SLD instead of strongly labelled data, with a novel unsupervised clustering stage. Experiments on 41 classes of sound events show that the proposed two-stage method trained on SLD achieves performance comparable to the previous state-of-the-art SED system trained on strongly labelled data, and is far better than another state-of-the-art SED system trained on weakly labelled data, which indicates the effectiveness of the proposed two-stage method trained on SLD without any onset/offset time of sound events.

Acoustic scene generation (ASG) is a task to generate waveforms for acoustic scenes. ASG can be used to generate audio scenes for movies and computer games. Recently, neural networks such as SampleRNN have been used for speech and music generation. However, ASG is more challenging due to its wide variety. In addition, evaluating a generative model is also difficult. In this paper, we propose to use a conditional SampleRNN model to generate acoustic scenes conditioned on the input classes. We also propose objective criteria to evaluate the quality and diversity of the generated samples based on classification accuracy. The experiments on the DCASE 2016 Task 1 acoustic scene data show that with the generated audio samples, a classification accuracy of 65.5% can be achieved compared to samples generated by a random model of 6.7% and samples from real recording of 83.1%. The performance of a classifier trained only on generated samples achieves an accuracy of 51.3%, as opposed to an accuracy of 6.7% with samples generated by a random model.

The goal of Acoustic Scene Classification (ASC) is to recognise the environment in which an audio waveform has been recorded. Recently, deep neural networks have been applied to ASC and have achieved state-of-the-art performance. However, few works have investigated how to visualise and understand what a neural network has learnt from acoustic scenes. Previous work applied local pooling after each convolutional layer, therefore reduced the size of the feature maps. In this paper, we suggest that local pooling is not necessary, but the size of the receptive field is important. We apply atrous Convolutional Neural Networks (CNNs) with global attention pooling as the classification model. The internal feature maps of the attention model can be visualised and explained. On the Detection and Classification of Acoustic Scenes and Events (DCASE) 2018 dataset, our proposed method achieves an accuracy of 72.7 %, significantly outperforming the CNNs without dilation at 60.4 %. Furthermore, our results demonstrate that the learnt feature maps contain rich information on acoustic scenes in the time-frequency domain.

In this paper, we address the problem of bird audio detection and propose a new convolutional neural network architecture together with a divergence based information channel weighing strategy in order to achieve improved state-of-the-art performance and faster convergence. The effectiveness of the methodology is shown on the Bird Audio Detection Challenge 2018 (Detection and Classification of Acoustic Scenes and Events Challenge, Task 3) development data set.

Single-channel signal separation and deconvolution aims to separate and deconvolve individual sources from a single-channel mixture. Single-channel signal separation and deconvolution is a challenging problem in which no prior knowledge of the mixing filters is available. Both individual sources and mixing filters need to be estimated. In addition, a mixture may contain non-stationary noise which is unseen in the training set. We propose a synthesizing-decomposition (S-D) approach to solve the single-channel separation and deconvolution problem. In synthesizing, a generative model for sources is built using a generative adversarial network (GAN). In decomposition, both mixing filters and sources are optimized to minimize the reconstruction error of the mixture. The proposed S-D approach achieves a peak-to-noise-ratio (PSNR) of 18.9 dB and 15.4 dB in image inpainting and completion, outperforming a baseline convolutional neural network PSNR of 15.3 dB and 12.2 dB, respectively and achieves a PSNR of 13.2 dB in source separation together with deconvolution, outperforming a convolutive non-negative matrix factorization (NMF) baseline of 10.1 dB.

Sound event detection (SED) aims to detect when and recognize what sound events happen in an audio clip. Many supervised SED algorithms rely on strongly labelled data that contains the onset and offset annotations of sound events. However, many audio tagging datasets are weakly labelled, that is, only the presence of the sound events is known, without knowing their onset and offset annotations. In this paper, we propose a time–frequency (T–F) segmentation framework trained on weakly labelled data to tackle the sound event detection and separation problem. In training, a segmentation mapping is applied on a T–F representation, such as log mel spectrogram of an audio clip to obtain T–F segmentation masks of sound events. The T–F segmentation masks can be used for separating the sound events from the background scenes in the T–F domain. Then, a classification mapping is applied on the T–F segmentation masks to estimate the presence probabilities of the sound events. We model the segmentation mapping using a convolutional neural network and the classification mapping using a global weighted rank pooling. In SED, predicted onset and offset times can be obtained from the T–F segmentation masks. As a byproduct, separated waveforms of sound events can be obtained from the T–F segmentation masks. We remixed the DCASE 2018 Task 1 acoustic scene data with the DCASE 2018 Task 2 sound events data. When mixing under 0 dB, the proposed method achieved F1 scores of 0.534, 0.398, and 0.167 in audio tagging, frame-wise SED and event-wise SED, outperforming the fully connected deep neural network baseline of 0.331, 0.237, and 0.120, respectively. In T–F segmentation, we achieved an F1 score of 0.218, where previous methods were not able to do T–F segmentation.

Audio tagging is the task of predicting the presence or absence of sound classes within an audio clip. Previous work in audio tagging focused on relatively small datasets limited to recognizing a small number of sound classes. We investigate audio tagging on AudioSet, which is a dataset consisting of over 2 million audio clips and 527 classes. AudioSet is weakly labelled, in that only the presence or absence of sound classes is known for each clip, whereas the onset and offset times are unknown. To address the weakly labelled audio tagging problem, we propose attention neural networks as a way to attend the most salient parts of an audio clip. We bridge the connection between attention neural networks and multiple instance learning (MIL) methods, and propose decision-level and feature-level attention neural networks for audio tagging. We investigate attention neural networks modeled by different functions, depths, and widths. Experiments on AudioSet show that the feature-level attention neural network achieves a state-of-the-art mean average precision of 0.369, outperforming the best MIL method of 0.317 and Google's deep neural network baseline of 0.314. In addition, we discover that the audio tagging performance on AudioSet-embedding features has a weak correlation with the number of training samples and the quality of labels of each sound class.

In this paper, we present a gated convolutional neural network and a temporal attention-based localization method for audio classification, which won the 1st place in the large-scale weakly supervised sound event detection task of Detection and Classification of Acoustic Scenes and Events (DCASE) 2017 challenge. The audio clips in this task, which are extracted from YouTube videos, are manually labelled with one or more audio tags, but without time stamps of the audio events, hence referred to as weakly labelled data. Two subtasks are defined in this challenge including audio tagging and sound event detection using this weakly labelled data. We propose a convolutional recurrent neural network (CRNN) with learnable gated linear units (GLUs) non-linearity applied on the log Mel spectrogram. In addition, we propose a temporal attention method along the frames to predict the locations of each audio event in a chunk from the weakly labelled data. The performances of our systems were ranked the 1st and the 2nd as a team in these two sub-tasks of DCASE 2017 challenge with F value 55.6% and Equal error 0.73, respectively.

This paper investigates the Audio Set classification. Audio Set is a large scale weakly labelled dataset (WLD) of audio clips. In WLD only the presence of a label is known, without knowing the happening time of the labels. We propose an attention model to solve this WLD problem and explain the attention model from a novel probabilistic perspective. Each audio clip in Audio Set consists of a collection of features. We call each feature as an instance and the collection as a bag following the terminology in multiple instance learning. In the attention model, each instance in the bag has a trainable probability measure for each class. The classification of the bag is the expectation of the classification output of the instances in the bag with respect to the learned probability measure. Experiments show that the proposed attention model achieves a mAP of 0.327 on Audio Set, outperforming the Google's baseline of 0.314.

Source separation (SS) aims to separate individual sources from an audio recording. Sound event detection (SED) aims to detect sound events from an audio recording. We propose a joint separation-classification (JSC) model trained only on weakly labelled audio data, that is, only the tags of an audio recording are known but the time of the events are unknown. First, we propose a separation mapping from the time-frequency (T-F) representation of an audio to the T-F segmentation masks of the audio events. Second, a classification mapping is built from each T-F segmentation mask to the presence probability of each audio event. In the source separation stage, sources of audio events and time of sound events can be obtained from the T-F segmentation masks. The proposed method achieves an equal error rate (EER) of 0.14 in SED, outperforming deep neural network baseline of 0.29. Source separation SDR of 8.08 dB is obtained by using global weighted rank pooling (GWRP) as probability mapping, out-performing the global max pooling (GMP) based probability mapping giving SDR at 0.03 dB. Source code of our work is published.

Audio tagging aims to assign one or several tags to an audio clip. Most of the datasets are weakly labelled, which means only the tags of the clip are known, without knowing the occurrence time of the tags. The labeling of an audio clip is often based on the audio events in the clip and no event level label is provided to the user. Previous works have used the bag of frames model assume the tags occur all the time, which is not the case in practice. We propose a joint detection-classification (JDC) model to detect and classify the audio clip simultaneously. The JDC model has the ability to attend to informative and ignore uninformative sounds. Then only informative regions are used for classification. Experimental results on the “CHiME Home” dataset show that the JDC model reduces the equal error rate (EER) from 19.0% to 16.9%. More interestingly, the audio event detector is trained successfully without needing the event level label.

Environmental audio tagging is a newly proposed task to predict the presence or absence of a specific audio event in a chunk. Deep neural network (DNN) based methods have been successfully adopted for predicting the audio tags in the domestic audio scene. In this paper, we propose to use a convolutional neural network (CNN) to extract robust features from mel-filter banks (MFBs), spectrograms or even raw waveforms for audio tagging. Gated recurrent unit (GRU) based recurrent neural networks (RNNs) are then cascaded to model the long-term temporal structure of the audio signal. To complement the input information, an auxiliary CNN is designed to learn on the spatial features of stereo recordings. We evaluate our proposed methods on Task 4 (audio tagging) of the Detection and Classification of Acoustic Scenes and Events 2016 (DCASE 2016) challenge. Compared with our recent DNN-based method, the proposed structure can reduce the equal error rate (EER) from 0.13 to 0.11 on the development set. The spatial features can further reduce the EER to 0.10. The performance of the end-to-end learning on raw waveforms is also comparable. Finally, on the evaluation set, we get the state-of-the-art performance with 0.12 EER while the performance of the best existing system is 0.15 EER.

Audio tagging aims to perform multi-label classification on audio \nchunks and it is a newly proposed task in the Detection and \nClassification of Acoustic Scenes and Events 2016 (DCASE \n2016) challenge. This task encourages research efforts to better \nanalyze and understand the content of the huge amounts of \naudio data on the web. The difficulty in audio tagging is that \nit only has a chunk-level label without a frame-level label. This \npaper presents a weakly supervised method to not only predict \nthe tags but also indicate the temporal locations of the occurred \nacoustic events. The attention scheme is found to be effective \nin identifying the important frames while ignoring the unrelated \nframes. The proposed framework is a deep convolutional recurrent \nmodel with two auxiliary modules: an attention module \nand a localization module. The proposed algorithm was evaluated \non the Task 4 of DCASE 2016 challenge. State-of-the-art \nperformance was achieved on the evaluation set with equal error \nrate (EER) reduced from 0.13 to 0.11, compared with the \nconvolutional recurrent baseline system.