English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Although great progresses have been made in automatic speech recognition(ASR), significant performance degradation is still observed when recognizingmulti-talker mixed speech. In this paper, we propose and evaluate severalarchitectures to address this problem under the assumption that only a singlechannel of mixed signal is available. Our technique extends permutationinvariant training (PIT) by introducing the front-end feature separation modulewith the minimum mean square error (MSE) criterion and the back-end recognitionmodule with the minimum cross entropy (CE) criterion. More specifically, duringtraining we compute the average MSE or CE over the whole utterance for eachpossible utterance-level output-target assignment, pick the one with theminimum MSE or CE, and optimize for that assignment. This strategy elegantlysolves the label permutation problem observed in the deep learning basedmulti-talker mixed speech separation and recognition systems. The proposedarchitectures are evaluated and compared on an artificially mixed AMI datasetwith both two- and three-talker mixed speech. The experimental results indicatethat our proposed architectures can cut the word error rate (WER) by 45.0% and25.0% relatively against the state-of-the-art single-talker speech recognitionsystem across all speakers when their energies are comparable, for two- andthree-talker mixed speech, respectively. To our knowledge, this is the firstwork on the multi-talker mixed speech recognition on the challengingspeaker-independent spontaneous large vocabulary continuous speech task.

Single-Channel Multi-talker Speech Recognition with Permutation Invariant Training