Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
97 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Speeding Up Permutation Invariant Training for Source Separation (2107.14445v1)

Published 30 Jul 2021 in eess.AS and cs.SD

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.

User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (5)
  1. Thilo von Neumann (16 papers)
  2. Christoph Boeddeker (36 papers)
  3. Keisuke Kinoshita (44 papers)
  4. Marc Delcroix (94 papers)
  5. Reinhold Haeb-Umbach (60 papers)
Citations (6)

Summary

We haven't generated a summary for this paper yet.