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

In this work we propose a novel token-based training strategy that improvesTransformer-Transducer (T-T) based speaker change detection (SCD) performance.The conventional T-T based SCD model loss optimizes all output tokens equally.Due to the sparsity of the speaker changes in the training data, theconventional T-T based SCD model loss leads to sub-optimal detection accuracy.To mitigate this issue, we use a customized edit-distance algorithm to estimatethe token-level SCD false accept (FA) and false reject (FR) rates duringtraining and optimize model parameters to minimize a weighted combination ofthe FA and FR, focusing the model on accurately predicting speaker changes. Wealso propose a set of evaluation metrics that align better with commercial usecases. Experiments on a group of challenging real-world datasets show that theproposed training method can significantly improve the overall performance ofthe SCD model with the same number of parameters.

Augmenting Transformer-Transducer Based Speaker Change Detection With Token-Level Training Loss