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

Discrete audio representation, aka audio tokenization, has seen renewedinterest driven by its potential to facilitate the application of text languagemodeling approaches in audio domain. To this end, various compression andrepresentation-learning based tokenization schemes have been proposed. However,there is limited investigation into the performance of compression-based audiotokens compared to well-established mel-spectrogram features across variousspeaker and speech related tasks. In this paper, we evaluate compression basedaudio tokens on three tasks: Speaker Verification, Diarization and(Multi-lingual) Speech Recognition. Our findings indicate that (i) the modelstrained on audio tokens perform competitively, on average within $1\%$ ofmel-spectrogram features for all the tasks considered, and do not surpass themyet. (ii) these models exhibit robustness for out-of-domain narrowband data,particularly in speaker tasks. (iii) audio tokens allow for compression to 20xcompared to mel-spectrogram features with minimal loss of performance in speechand speaker related tasks, which is crucial for low bit-rate applications, and(iv) the examined Residual Vector Quantization (RVQ) based audio tokenizerexhibits a low-pass frequency response characteristic, offering a plausibleexplanation for the observed results, and providing insight for futuretokenizer designs.

Discrete Audio Representation as an Alternative to Mel-Spectrograms for Speaker and Speech Recognition