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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Majority of speech signals across different scenarios are never availablewith well-defined audio segments containing only a single speaker. A typicalconversation between two speakers consists of segments where their voicesoverlap, interrupt each other or halt their speech in between multiplesentences. Recent advancements in diarization technology leverage neuralnetwork-based approaches to improvise multiple subsystems of speakerdiarization system comprising of extracting segment-wise embedding features anddetecting changes in the speaker during conversation. However, to identifyspeaker through clustering, models depend on methodologies like PLDA togenerate similarity measure between two extracted segments from a givenconversational audio. Since these algorithms ignore the temporal structure ofconversations, they tend to achieve a higher Diarization Error Rate (DER), thusleading to misdetections both in terms of speaker and change identification.Therefore, to compare similarity of two speech segments both independently andsequentially, we propose a Bi-directional Long Short-term Memory network forestimating the elements present in the similarity matrix. Once the similaritymatrix is generated, Agglomerative Hierarchical Clustering (AHC) is applied tofurther identify speaker segments based on thresholding. To evaluate theperformance, Diarization Error Rate (DER%) metric is used. The proposed modelachieves a low DER of 34.80% on a test set of audio samples derived from ICSIMeeting Corpus as compared to traditional PLDA based similarity measurementmechanism which achieved a DER of 39.90%.

Bi-LSTM Scoring Based Similarity Measurement with Agglomerative Hierarchical Clustering (AHC) for Speaker Diarization