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

We study the individuality of the human voice with respect to a widely usedfeature representation of speech utterances, namely, the i-vector model. As afirst step toward this goal, we compare and contrast uniqueness measuresproposed for different biometric modalities. Then, we introduce a newuniqueness measure that evaluates the entropy of i-vectors while taking intoaccount speaker level variations. Our measure operates in the discrete featurespace and relies on accurate estimation of the distribution of i-vectors.Therefore, i-vectors are quantized while ensuring that both the quantized andoriginal representations yield similar speaker verification performance.Uniqueness estimates are obtained from two newly generated datasets and thepublic VoxCeleb dataset. The first custom dataset contains more than one and ahalf million speech samples of 20,741 speakers obtained from TEDx Talks videos.The second one includes over twenty one thousand speech samples from 1,595actors that are extracted from movie dialogues. Using this data, we analyzedhow several factors, such as the number of speakers, number of samples perspeaker, sample durations, and diversity of utterances affect uniquenessestimates. Most notably, we determine that the discretization of i-vectors doesnot cause a reduction in speaker recognition performance. Our results show thatthe degree of distinctiveness offered by i-vector-based representation mayreach 43-70 bits considering 5-second long speech samples; however, under lessconstrained variations in speech, uniqueness estimates are found to reduce byaround 30 bits. We also find that doubling the sample duration increases thedistinctiveness of the i-vector representation by around 20 bits.

Estimating Uniqueness of I-Vector Representation of Human Voice