Reverberation as Supervision for Speech Separation

11/15/2022
by   Rohith Aralikatti, et al.
0

This paper proposes reverberation as supervision (RAS), a novel unsupervised loss function for single-channel reverberant speech separation. Prior methods for unsupervised separation required the synthesis of mixtures of mixtures or assumed the existence of a teacher model, making them difficult to consider as potential methods explaining the emergence of separation abilities in an animal's auditory system. We assume the availability of two-channel mixtures at training time, and train a neural network to separate the sources given one of the channels as input such that the other channel may be predicted from the separated sources. As the relationship between the room impulse responses (RIRs) of each channel depends on the locations of the sources, which are unknown to the network, the network cannot rely on learning that relationship. Instead, our proposed loss function fits each of the separated sources to the mixture in the target channel via Wiener filtering, and compares the resulting mixture to the ground-truth one. We show that minimizing the scale-invariant signal-to-distortion ratio (SI-SDR) of the predicted right-channel mixture with respect to the ground truth implicitly guides the network towards separating the left-channel sources. On a semi-supervised reverberant speech separation task based on the WHAMR! dataset, using training data where just 5 10 70 supervision on the full training set, while a model trained only on the labeled data obtains 43

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/23/2020

Unsupervised Sound Separation Using Mixtures of Mixtures

In recent years, rapid progress has been made on the problem of single-c...
research
06/15/2021

Teacher-Student MixIT for Unsupervised and Semi-supervised Speech Separation

In this paper, we introduce a novel semi-supervised learning framework f...
research
12/17/2021

Discretization and Re-synthesis: an alternative method to solve the Cocktail Party Problem

Deep learning based models have significantly improved the performance o...
research
10/23/2020

Training Noisy Single-Channel Speech Separation With Noisy Oracle Sources: A Large Gap and A Small Step

As the performance of single-channel speech separation systems has impro...
research
02/08/2022

Unsupervised Source Separation via Self-Supervised Training

We introduce two novel unsupervised (blind) source separation methods, w...
research
10/20/2021

Adapting Speech Separation to Real-World Meetings Using Mixture Invariant Training

The recently-proposed mixture invariant training (MixIT) is an unsupervi...
research
06/01/2021

Sparse, Efficient, and Semantic Mixture Invariant Training: Taming In-the-Wild Unsupervised Sound Separation

Supervised neural network training has led to significant progress on si...

Please sign up or login with your details

Forgot password? Click here to reset