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

Pre-trained model representations have demonstrated state-of-the-artperformance in speech recognition, natural language processing, and otherapplications. Speech models, such as Bidirectional Encoder Representations fromTransformers (BERT) and Hidden units BERT (HuBERT), have enabled generatinglexical and acoustic representations to benefit speech recognitionapplications. We investigated the use of pre-trained model representations forestimating dimensional emotions, such as activation, valence, and dominance,from speech. We observed that while valence may rely heavily on lexicalrepresentations, activation and dominance rely mostly on acoustic information.In this work, we used multi-modal fusion representations from pre-trainedmodels to generate state-of-the-art speech emotion estimation, and we showed a100% and 30% relative improvement in concordance correlation coefficient (CCC)on valence estimation compared to standard acoustic and lexical baselines.Finally, we investigated the robustness of pre-trained model representationsagainst noise and reverberation degradation and noticed that lexical andacoustic representations are impacted differently. We discovered that lexicalrepresentations are more robust to distortions compared to acousticrepresentations, and demonstrated that knowledge distillation from amulti-modal model helps to improve the noise-robustness of acoustic-basedmodels.

Pre-trained Model Representations and their Robustness against Noise for Speech Emotion Analysis