Low latency models are critical for real-time speech enhancement applications, such as hearing aids and hearables. However, the sub-millisecond latency space for resource-constrained hearables remains underexplored. We demonstrate speech enhancement using a computationally efficient minimum-phase FIR filter, enabling sample-by-sample processing to achieve mean algorithmic latency of 0.32 ms to 1.25 ms. With a single microphone, we observe a mean SI-SDRi of 4.1 dB. The approach shows generalization with a DNSMOS increase of 0.2 on unseen audio recordings. We use a lightweight LSTM-based model of 626k parameters to generate FIR taps. Using a real hardware implementation on a low-power DSP, our system can run with 376 MIPS and a mean end-to-end latency of 3.35 ms. In addition, we provide a comparison with existing low-latency spectral masking techniques. We hope this work will enable a better understanding of latency and can be used to improve the comfort and usability of hearables.

Supervised models for speech enhancement are trained using artificially generated mixtures of clean speech and noise signals. However, the synthetic training conditions may not accurately reflect real-world conditions encountered during testing. This discrepancy can result in poor performance when the test domain significantly differs from the synthetic training domain. To tackle this issue, the UDASE task of the 7th CHiME challenge aimed to leverage real-world noisy speech recordings from the test domain for unsupervised domain adaptation of speech enhancement models. Specifically, this test domain corresponds to the CHiME-5 dataset, characterized by real multi-speaker and conversational speech recordings made in noisy and reverberant domestic environments, for which ground-truth clean speech signals are not available. In this paper, we present the objective and subjective evaluations of the systems that were submitted to the CHiME-7 UDASE task, and we provide an analysis of the results. This analysis reveals a limited correlation between subjective ratings and several supervised nonintrusive performance metrics recently proposed for speech enhancement. Conversely, the results suggest that more traditional intrusive objective metrics can be used for in-domain performance evaluation using the reverberant LibriCHiME-5 dataset developed for the challenge. The subjective evaluation indicates that all systems successfully reduced the background noise, but always at the expense of increased distortion. Out of the four speech enhancement methods evaluated subjectively, only one demonstrated an improvement in overall quality compared to the unprocessed noisy speech, highlighting the difficulty of the task. The tools and audio material created for the CHiME-7 UDASE task are shared with the community.

A key challenge in machine learning is to generalize from training data to an application domain of interest. This work extends the recently-proposed mixture invariant training (MixIT) algorithm to perform unsupervised learning in the multi-channel setting. We use MixIT to train a model on far-field microphone array recordings of overlapping reverberant and noisy speech from the AMI Corpus. The models are trained on both supervised and unsupervised training data, and are tested on real AMI recordings containing overlapping speech. To objectively evaluate our models, we also use a synthetic multi-channel AMI test set. Holding network architectures constant, we find that semi-supervised fine-tuning of a model pretrained on a large and diverse single-channel dataset yields the largest improvement to SI-SNR and to human listening ratings across synthetic and real datasets, outperforming supervised models trained on well-matched synthetic data. Our results demonstrate that unsupervised learning through MixIT enables model adaptation on both single-and multi-channel real-world speech recordings.

This paper addresses the problem of species classification in bird song recordings. The massive amount of available field recordings of birds presents an opportunity to use machine learning to automatically track bird populations. However, it also poses a problem: such field recordings typically contain significant environmental noise and overlapping vocalizations that interfere with classification. The widely available training datasets for species identification also typically leave background species unlabeled. This leads classifiers to ignore vocalizations with a low signal-to-noise ratio. However, recent advances in unsupervised sound separation, such as mixture invariant training (MixIT), enable high quality separation of bird songs to be learned from such noisy recordings. In this paper, we demonstrate improved separation quality when training a MixIT model specifically for birdsong data, outperforming a general audio separation model by over 5 dB in SI-SNR improvement of reconstructed mixtures. We also demonstrate precision improvements with a downstream multi-species bird classifier across three independent datasets. The best classifier performance is achieved by taking the maximum model activations over the separated channels and original audio. Finally, we document additional classifier improvements, including taxonomic classification, augmentation by random low-pass filters, and additional channel normalization.

The recently-proposed mixture invariant training (MixIT) is an unsupervised method for training single-channel sound separation models because it does not require ground-truth isolated reference sources. In this paper, we investigate using MixIT to adapt a separation model on real far-field overlapping reverberant and noisy speech data from the AMI Corpus. The models are tested on real AMI recordings containing overlapping speech, and are evaluated subjectively by human listeners. To objectively evaluate our models, we also devise a synthetic AMI test set. For human evaluations on real recordings, we also propose a modification of the standard MUSHRA protocol to handle imperfect reference signals, which we call MUSHIRA. Holding network architectures constant, we find that a fine-tuned semi-supervised model yields the largest SI-SNR improvement, PESQ scores, and human listening ratings across synthetic and real datasets, outperforming unadapted generalist models trained on orders of magnitude more data. Our results show that unsupervised learning through MixIT enables model adaptation on real-world unlabeled spontaneous speech recordings.

Typically, neural network-based speech dereverberation models are trained on paired data, composed of a dry utterance and its corresponding reverberant utterance.The main limitation of this approach is that such models can only be trained on large amounts of data and a variety of room impulse responses when the data is synthetically reverberated, since acquiring real paired data is costly.In this paper we propose a CycleGAN-based approach that enables dereverberation models to be trained on unpaired data.We quantify the impact of using unpaired data by comparing the proposed unpaired model to a paired model with the same architecture and trained on the paired version of the same dataset.We show that the performance of the unpaired model is comparable to the performance of the paired model on two different datasets, according to objective evaluation metrics.Furthermore, we run two subjective evaluations and show that both models achieve comparable subjective quality on the AMI dataset, which was not seen during training.

We propose the novel task of distance-based sound separation, where sounds are separated based only on their distance from a single microphone.In the context of assisted listening devices, proximity provides a simple criterion for sound selection in noisy environments that would allow the user to focus on sounds relevant to a local conversation.We demonstrate the feasibility of this approach by training a neural network to separate near sounds from far sounds in single channel synthetic reverberant mixtures, relative to a threshold distance defining the boundary between near and far.With a single nearby speaker and four distant speakers, the model improves scale-invariant signal to noise ratio by 4.4 dB for near sounds and 6.8 dB for far sounds.

We present an end-to-end deep network model that performs meeting diarization from single-channel audio recordings. End-to-end diarization models have the advantage of handling speaker overlap and enabling straightforward handling of discriminative training, unlike traditional clustering-based diarization methods. The proposed system is designed to handle meetings with unknown numbers of speakers, using variable-number permutation-invariant cross-entropy based loss functions. We introduce several components that appear to help with diarization performance, including a local convolutional network followed by a global self-attention module, multitask transfer learning using a speaker identification component, and a sequential approach where the model is refined with a second stage. These are trained and validated on simulated meeting data based on LibriSpeech and LibriTTS datasets; final evaluations are done using LibriCSS, which consists of simulated meetings recorded using real acoustics via loudspeaker playback. The proposed model performs better than previously proposed end-to-end diarization models on these data.

We propose a benchmark of state-of-the-art sound event detection systems (SED). We design synthetic evaluation sets to focus on specific sound event detection challenges. We analyze the performance of the submissions to DCASE 2020 Task 4 as a function of time-related modifications (time position of an event and length of clips) and study the impact of non-target sound events and reverberation. We show that temporal localization of sound events remains a challenge for SED systems. We also show that reverberation and non-target sound events severely degrade system performance. In the latter case, sound separation seems like a promising solution.

We introduce the Free Universal Sound Separation (FUSS) dataset, a new corpus for experiments in separating mixtures of an unknown number of sounds from an open domain of sound types. The dataset consists of 23 hours of single-source audio data drawn from 357 classes, which are used to create mixtures of one to four sources. To simulate reverberation, an acoustic room simulator is used to generate impulse responses of box-shaped rooms with frequency-dependent reflective walls. Additional open-source data augmentation tools are also provided to produce new mixtures with different combinations of sources and room simulations. Finally, we introduce an open-source baseline separation model, based on an improved time-domain convolutional network (TDCN++), that can separate a variable number of sources in a mixture. This model achieves 9.8 dB of scale-invariant signal-to-noise ratio improvement (SI-SNRi) on mixtures with two to four sources, while reconstructing single-source inputs with 35.8 dB absolute SI-SNR. We hope this dataset will lower the barrier to new research and allow for fast iteration and application of novel techniques from other machine learning domains to the sound separation challenge.

Multi-speaker speech recognition of unsegmented recordings has diverse applications such as meeting transcription and automatic subtitle generation. With technical advances in systems dealing with speech separation, speaker diarization, and automatic speech recognition (ASR) in the last decade, it has become possible to build pipelines that achieve reasonable error rates on this task. In this paper, we propose an end-to-end modular system for the LibriCSS meeting data, which combines independently trained separation, diarization, and recognition components, in that order. We study the effect of different state-of-the-art methods at each stage of the pipeline, and report results using task-specific metrics like SDR and DER, as well as downstream WER. Experiments indicate that the problem of overlapping speech for diarization and ASR can be effectively mitigated with the presence of a well-trained separation module. Our best system achieves a speaker-attributed WER of 12.7%, which is close to that of a non-overlapping ASR.

This work introduces sequential neural beamforming, which alternates between neural network based spectral separation and beamforming based spatial separation. Our neural networks for separation use an advanced convolutional architecture trained with a novel stabilized signal-to-noise ratio loss function. For beamforming, we explore multiple ways of computing time-varying covariance matrices, including factorizing the spatial covariance into a time-varying amplitude component and a time-invariant spatial component, as well as using block-based techniques. In addition, we introduce a multi-frame beamforming method which improves the results significantly by adding contextual frames to the beamforming formulations. We extensively evaluate and analyze the effects of window size, block size, and multi-frame context size for these methods. Our best method utilizes a sequence of three neural separation and multi-frame time-invariant spatial beamforming stages, and demonstrates an average improvement of 2.75 dB in scale-invariant signal-to-noise ratio and 14.2% absolute reduction in a comparative speech recognition metric across four challenging reverberant speech enhancement and separation tasks. We also use our three-speaker separation model to separate real recordings in the LibriCSS evaluation set into non-overlapping tracks, and achieve a better word error rate as compared to a baseline mask based beamformer.

Time-domain audio separation networks based on dilated temporal convolutions have recently been shown to perform very well compared to methods that are based on a time-frequency representation in speech separation tasks, even outperforming an oracle binary time-frequency mask of the speakers. This paper investigates the performance of such a time-domain network (Conv-TasNet) for speech denoising in a real-time setting, comparing various parameter settings. Most importantly, different amounts of lookahead are evaluated and compared to the baseline of a fully causal model. We show that a large part of the increase in performance between a causal and non-causal model is achieved with a lookahead of only 20 milliseconds, demonstrating the usefulness of even small lookaheads for many real-time applications.

Deep learning approaches have recently achieved impressive performance on both audio source separation and sound classification. Most audio source separation approaches focus only on separating sources belonging to a restricted domain of source classes, such as speech and music. However, recent work has demonstrated the possibility of "universal sound separation", which aims to separate acoustic sources from an open domain, regardless of their class. In this paper, we utilize the semantic information learned by sound classifier networks trained on a vast amount of diverse sounds to improve universal sound separation. In particular, we show that semantic embeddings extracted from a sound classifier can be used to condition a separation network, providing it with useful additional information. This approach is especially useful in an iterative setup, where source estimates from an initial separation stage and their corresponding classifier-derived embeddings are fed to a second separation network. By performing a thorough hyperparameter search consisting of over a thousand experiments, we find that classifier embeddings from oracle clean sources provide nearly one dB of SNR gain, and our best iterative models achieve a significant fraction of this oracle performance, establishing a new state-of-the-art for universal sound separation.

In recent years, rapid progress has been made on the problem of single-channel sound separation using supervised training of deep neural networks. In such supervised approaches, a model is trained to predict the component sources from synthetic mixtures created by adding up isolated ground-truth sources. Reliance on this synthetic training data is problematic because good performance depends upon the degree of match between the training data and real-world audio, especially in terms of the acoustic conditions and distribution of sources. The acoustic properties can be challenging to accurately simulate, and the distribution of sound types may be hard to replicate. In this paper, we propose a completely unsupervised method, mixture invariant training (MixIT), that requires only single-channel acoustic mixtures. In MixIT, training examples are constructed by mixing together existing mixtures, and the model separates them into a variable number of latent sources, such that the separated sources can be remixed to approximate the original mixtures. We show that MixIT can achieve competitive performance compared to supervised methods on speech separation. Using MixIT in a semi-supervised learning setting enables unsupervised domain adaptation and learning from large amounts of real-world data without ground-truth source waveforms. In particular, we significantly improve reverberant speech separation performance by incorporating reverberant mixtures, train a speech enhancement system from noisy mixtures, and improve universal sound separation by incorporating a large amount of in-the-wild data.

In speech enhancement and source separation, signal-to-noise ratio is a ubiquitous objective measure of denoising/separation quality. A decade ago, the BSS_eval toolkit was developed to give researchers worldwide a way to evaluate the quality of their algorithms in a simple, fair, and hopefully insightful way: it attempted to account for channel variations, and to not only evaluate the total distortion in the estimated signal but also split it in terms of various factors such as remaining interference, newly added artifacts, and channel errors. In recent years, hundreds of papers have been relying on this toolkit to evaluate their proposed methods and compare them to previous works, often arguing that differences on the order of 0.1 dB proved the effectiveness of a method over others. We argue here that the signal-to-distortion ratio (SDR) implemented in the BSS_eval toolkit has generally been improperly used and abused, especially in the case of single-channel separation, resulting in misleading results. We propose to use a slightly modified definition, resulting in a simpler, more robust measure, called scale-invariant SDR (SI-SDR). We present various examples of critical failure of the original SDR that SI-SDR overcomes.

In recent years, deep networks have led to dramatic improvements in speech enhancement by framing it as a data-driven pattern recognition problem. In many modern enhancement systems, large amounts of data are used to train a deep network to estimate masks for complex-valued short-time Fourier transforms (STFTs) to suppress noise and preserve speech. However, current masking approaches often neglect two important constraints: STFT consistency and mixture consistency. Without STFT consistency, the system's output is not necessarily the STFT of a time-domain signal, and without mixture consistency, the sum of the estimated sources does not necessarily equal the input mixture. Furthermore, the only previous approaches that apply mixture consistency use real-valued masks; mixture consistency has been ignored for complex-valued masks.

Historically, sparse methods and neural networks, particularly modern deep learning methods, have been relatively disparate areas. Sparse methods are typically used for signal enhancement, compression, and recovery, usually in an unsupervised framework, while neural networks commonly rely on a supervised training set. In this paper, we use the specific problem of sequential sparse recovery, which models a sequence of observations over time using a sequence of sparse coefficients, to show how algorithms for sparse modeling can be combined with supervised deep learning to improve sparse recovery. Specifically, we show that the iterative soft-thresholding algorithm (ISTA) for sequential sparse recovery corresponds to a stacked recurrent neural network (RNN) under specific architecture and parameter constraints. Then we demonstrate the benefit of training this RNN with backpropagation using supervised data for the task of column-wise compressive sensing of images. This training corresponds to adaptation of the original iterative thresholding algorithm and its parameters. Thus, we show by example that sparse modeling can provide a rich source of principled and structured deep network architectures that can be trained to improve performance on specific tasks.

Deep unfolding has recently been proposed to derive novel deep network architectures from model-based approaches. In this paper, we consider its application to multichannel source separation. We unfold a multichannel Gaussian mixture model (MCGMM), resulting in a deep MCGMM computational network that directly processes complex-valued frequency-domain multichannel audio and has an architecture defined explicitly by a generative model, thus combining the advantages of deep networks and model-based approaches. We further extend the deep MCGMM by modeling the GMM states using an MRF, whose unfolded mean-field inference updates add dynamics across layers. Experiments on source separation for multichannel mixtures of two simultaneous speakers shows that the deep MCGMM leads to improved performance with respect to the original MCGMM model.

Recurrent neural networks are powerful models for processing sequential data, but they are generally plagued by vanishing and exploding gradient problems. Unitary recurrent neural networks (uRNNs), which use unitary recurrence matrices, have recently been proposed as a means to avoid these issues. However, in previous experiments, the recurrence matrices were restricted to be a product of parameterized unitary matrices, and an open question remains: when does such a parameterization fail to represent all unitary matrices, and how does this restricted representational capacity limit what can be learned? To address this question, we propose full-capacity uRNNs that optimize their recurrence matrix over all unitary matrices, leading to significantly improved performance over uRNNs that use a restricted-capacity recurrence matrix. Our contribution consists of two main components. First, we provide a theoretical argument to determine if a unitary parameterization has restricted capacity. Using this argument, we show that a recently proposed unitary parameterization has restricted capacity for hidden state dimension greater than 7. Second,we show how a complete, full-capacity unitary recurrence matrix can be optimized over the differentiable manifold of unitary matrices. The resulting multiplicative gradient step is very simple and does not require gradient clipping or learning rate adaptation. We confirm the utility of our claims by empirically evaluating our new full-capacity uRNNs on both synthetic and natural data, achieving superior performance compared to both LSTMs and the original restricted-capacity uRNNs.

Many voice activity detection (VAD) systems use the magnitude of complex-valued spectral representations. However, using only the magnitude often does not fully characterize the statistical behavior of the complex values. We present two novel methods for performing VAD on single- and dual-channel audio that do completely account for the second-order statistical behavior of complex data. Our methods exploit the second-order noncircularity (also known as impropriety) of complex subbands of speech and noise. Since speech tends to be more improper than noise, higher impropriety suggests speech activity. Our single-channel method is blind in the sense that it is unsupervised and, unlike many VAD systems, does not rely on non-speech periods for noise parameter estimation. Our methods achieve improved performance over other state-of-the-art magnitude-based VADs on the QUT-NOISE-TIMIT corpus, which indicates that impropriety is a compelling new feature for voice activity detection.

The strong uncorrelating transform (SUT) provides estimates of independent components from linear mixtures using only second-order information, provided that the components have unique circularity coefficients. We propose a processing framework for generating complex-valued subbands from real-valued mixtures of speech and noise where the objective is to control the likely values of the sample circularity coefficients of the underlying speech and noise components in each subband. We show how several processing parameters affect the noncircularity of speech-like and noise components in the subband, ultimately informing parameter choices that allow for estimation of each of the components in a subband using the SUT. Additionally, because the speech and noise components will have unique sample circularity coefficients, this statistic can be used to identify time-frequency regions that contain voiced speech. We give an example of the recovery of the circularity coefficients of a real speech signal from a two-channel noisy mixture at -25 dB SNR, which demonstrates how the estimates of noncircularity can reveal the time-frequency structure of a speech signal in very high levels of noise. Finally, we present the results of a voice activity detection (VAD) experiment showing that two new circularity-based statistics, one of which is derived from the SUT processing, can achieve improved performance over state-of-the-art VADs in real-world recordings of noise.

In this paper, we relax a commonly-used assumption about a class of nonstationary random processes composed of modulated wide-sense stationary random processes: that the fundamental frequency of the modulator is stationary within the analysis window. To compensate for the relaxation of this assumption, we define the generalized DEMON (“demodulated noise”) spectrum representing modulation frequency, which we use to increase the coherence time of such signals. Increased coherence time means longer analysis windows, which provides higher SNR estimators. We use the example of detection on both synthetic and real-world passive sonar signals to demonstrate this increase.