The field of speech processing has undergone a transformative shift with the advent of deep
learning. The use of multiple processing layers has enabled the creation of models capable …

Speaker diarization is a task to label audio or video recordings with classes that correspond
to speaker identity, or in short, a task to identify “who spoke when”. In the early years …

Recurrent Neural Networks (RNNs) have long been the dominant architecture in sequence-
to-sequence learning. RNNs, however, are inherently sequential models that do not allow …

Recent studies in deep learning-based speech separation have proven the superiority of
time-domain approaches to conventional time-frequency-based methods. Unlike the time …

In this study, we present recent developments on ESPnet: End-to-End Speech Processing
toolkit, which mainly involves a recently proposed architecture called Conformer …

The dominant speech separation models are based on complex recurrent or convolution
neural network that model speech sequences indirectly conditioning on context, such as …

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 …

Single-channel, speaker-independent speech separation methods have recently seen great
progress. However, the accuracy, latency, and computational cost of such methods remain …

Speech enhancement and speech separation are two related tasks, whose purpose is to
extract either one or more target speech signals, respectively, from a mixture of sounds …

Speech separation is the task of separating target speech from background interference.
Traditionally, speech separation is studied as a signal processing problem. A more recent …