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

Recently, data-driven based Automatic Speech Recognition (ASR) systems haveachieved state-of-the-art results. And transfer learning is often used whenthose existing systems are adapted to the target domain, e.g., fine-tuning,retraining. However, in the processes, the system parameters may well deviatetoo much from the previously learned parameters. Thus, it is difficult for thesystem training process to learn knowledge from target domains meanwhile notforgetting knowledge from the previous learning process, which is called ascatastrophic forgetting (CF). In this paper, we attempt to solve the CF problemwith the lifelong learning and propose a novel multi-task learning (MTL)training framework for ASR. It considers reserving original knowledge andlearning new knowledge as two independent tasks, respectively. On the one hand,we constrain the new parameters not to deviate too far from the originalparameters and punish the new system when forgetting original knowledge. On theother hand, we force the new system to solve new knowledge quickly. Then, a MTLmechanism is employed to get the balance between the two tasks. We applied ourmethod to an End2End ASR task and obtained the best performance in both targetand original datasets.

A Multi-Task Learning Framework for Overcoming the Catastrophic Forgetting in Automatic Speech Recognition