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

Text language models have shown remarkable zero-shot capability ingeneralizing to unseen tasks when provided with well-formulated instructions.However, existing studies in speech processing primarily focus on limited orspecific tasks. Moreover, the lack of standardized benchmarks hinders a faircomparison across different approaches. Thus, we present Dynamic-SUPERB, abenchmark designed for building universal speech models capable of leveraginginstruction tuning to perform multiple tasks in a zero-shot fashion. To achievecomprehensive coverage of diverse speech tasks and harness instruction tuning,we invite the community to collaborate and contribute, facilitating the dynamicgrowth of the benchmark. To initiate, Dynamic-SUPERB features 55 evaluationinstances by combining 33 tasks and 22 datasets. This spans a broad spectrum ofdimensions, providing a comprehensive platform for evaluation. Additionally, wepropose several approaches to establish benchmark baselines. These include theutilization of speech models, text language models, and the multimodal encoder.Evaluation results indicate that while these baselines perform reasonably onseen tasks, they struggle with unseen ones. We also conducted an ablation studyto assess the robustness and seek improvements in the performance. We releaseall materials to the public and welcome researchers to collaborate on theproject, advancing technologies in the field together.

Dynamic-SUPERB: Towards A Dynamic, Collaborative, and Comprehensive Instruction-Tuning Benchmark for Speech