Related papers: MixCycle: Unsupervised Speech Separation via Cycli…
Permutation-invariant training (PIT) is a dominant approach for addressing the permutation ambiguity problem in talker-independent speaker separation. Leveraging spatial information afforded by microphone arrays, we propose a new training…
In natural language processing and vision, pretraining is utilized to learn effective representations. Unfortunately, the success of pretraining does not easily carry over to time series due to potential mismatch between sources and target.…
This paper presents a joint source separation algorithm that simultaneously reduces acoustic echo, reverberation and interfering sources. Target speeches are separated from the mixture by maximizing independence with respect to the other…
The recently advanced unsupervised learning approaches use the siamese-like framework to compare two "views" from the same image for learning representations. Making the two views distinctive is a core to guarantee that unsupervised methods…
In this paper, we study whether music source separation can be used as a pre-training strategy for music representation learning, targeted at music classification tasks. To this end, we first pre-train U-Net networks under various music…
Consider a time series of measurements of the state of an evolving system, x(t), where x has two or more components. This paper shows how to perform nonlinear blind source separation; i.e., how to determine if these signals are equal to…
We consider the problem of audio voice separation for binaural applications, such as earphones and hearing aids. While today's neural networks perform remarkably well (separating $4+$ sources with 2 microphones) they assume a known or fixed…
Strong semantic representations improve the convergence and generation quality of diffusion and flow models. Existing approaches largely rely on external models, which require separate training, operate on misaligned objectives, and exhibit…
The separation of single-channel underwater acoustic signals is a challenging problem with practical significance. Few existing studies focus on the source separation problem with unknown numbers of signals, and how to evaluate the…
Speech separation has been well developed, with the very successful permutation invariant training (PIT) approach, although the frequent label assignment switching happening during PIT training remains to be a problem when better…
Binaural speech separation in real-world scenarios often involves moving speakers. Most current speech separation methods use utterance-level permutation invariant training (u-PIT) for training. In inference time, however, the order of…
We propose a training scheme to train neural network-based source separation algorithms from scratch when parallel clean data is unavailable. In particular, we demonstrate that an unsupervised spatial clustering algorithm is sufficient to…
Speech separation with several speakers is a challenging task because of the non-stationarity of the speech and the strong signal similarity between interferent sources. Current state-of-the-art solutions can separate well the different…
Recent progress in deep learning has enabled many advances in sound separation and visual scene understanding. However, extracting sound sources which are apparent in natural videos remains an open problem. In this work, we present…
While pre-trained language models have achieved great success on various natural language understanding tasks, how to effectively leverage them into non-autoregressive generation tasks remains a challenge. To solve this problem, we present…
Building a good speech recognition system usually requires large amounts of transcribed data, which is expensive to collect. To tackle this problem, many unsupervised pre-training methods have been proposed. Among these methods, Masked…
Music source separation is focused on extracting distinct sonic elements from composite tracks. Historically, many methods have been grounded in supervised learning, necessitating labeled data, which is occasionally constrained in its…
Speech separation has been studied in time domain because of lower latency and higher performance compared to time-frequency domain. The masking-based method has been mostly used in time domain, and the other common method (mapping-based)…
We consider the problem of separating a particular sound source from a single-channel mixture, based on only a short sample of the target source. Using SoundFilter, a wave-to-wave neural network architecture, we can train a model without…
Self-supervised learning leverages unlabeled data effectively, improving label efficiency and generalization to domains without labeled data. While recent work has studied generalization to more acoustic/linguistic domains, languages, and…