SDCLJun 22, 2017

Speaker Recognition with Cough, Laugh and "Wei"

arXiv:1706.07860v115 citations
Originality Synthesis-oriented
AI Analysis

This addresses the problem of speaker recognition in scenarios with disguised or short speech for applications like security and authentication, though it is incremental as it applies existing deep learning methods to new types of data.

The paper tackles speaker recognition using short, trivial speech events like coughs, laughs, and a Chinese 'Wei', finding that these events contain rich speaker information despite their brevity and spectral challenges. With a deep feature learning approach, they achieved equal error rates (EER) of 10%-14% for events lasting 0.2-1.0 seconds.

This paper proposes a speaker recognition (SRE) task with trivial speech events, such as cough and laugh. These trivial events are ubiquitous in conversations and less subjected to intentional change, therefore offering valuable particularities to discover the genuine speaker from disguised speech. However, trivial events are often short and idiocratic in spectral patterns, making SRE extremely difficult. Fortunately, we found a very powerful deep feature learning structure that can extract highly speaker-sensitive features. By employing this tool, we studied the SRE performance on three types of trivial events: cough, laugh and "Wei" (a short Chinese "Hello"). The results show that there is rich speaker information within these trivial events, even for cough that is intuitively less speaker distinguishable. With the deep feature approach, the EER can reach 10%-14% with the three trivial events, despite their extremely short durations (0.2-1.0 seconds).

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes