Speeding Up Permutation Invariant Training for Source Separation
arXiv:2107.14445
Abstract
Permutation invariant training (PIT) is a widely used training criterion for neural network-based source separation, used for both utterance-level separation with utterance-level PIT (uPIT) and separation of long recordings with the recently proposed Graph-PIT. When implemented naively, both suffer from an exponential complexity in the number of utterances to separate, rendering them unusable for large numbers of speakers or long realistic recordings. We present a decomposition of the PIT criterion into the computation of a matrix and a strictly monotonously increasing function so that the permutation or assignment problem can be solved efficiently with several search algorithms. The Hungarian algorithm can be used for uPIT and we introduce various algorithms for the Graph-PIT assignment problem to reduce the complexity to be polynomial in the number of utterances.
arXiv commentAccepted at 14th ITG Conference on Speech Communication
References in corpus (4)
- Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation
- Recognizing Overlapped Speech in Meetings: A Multichannel Separation Approach Using Neural Networks
- CHiME-6 Challenge:Tackling Multispeaker Speech Recognition for Unsegmented Recordings
- Graph-PIT: Generalized permutation invariant training for continuous separation of arbitrary numbers of speakers