Automatic Speaker Recognition using Phase based Features: Developing FM based Automatic Speaker Recognition System to Complement Conventional Systems - Julien Epps,Tharmarajah Thiruvaran,Eliathamby Ambikairajah
-30% ar kodu BOOKS
Piegāde 12-18 darba dienu laikā
30 dienu atgriešanas politika
A theory has been developed to show that the FM component can be extracted using second-order all pole modelling, and a technique for extracting FM features using this model is proposed, to produce very smooth, slowly varying FM features that are effective for speaker recognition. This is evaluated on NIST 2008 international speaker recognition evaluation. Optimum Frequency band allocation specifically to c ... Pilns apraksts
Jums varētu patikt arī
Aprašymas
A theory has been developed to show that the FM component can be extracted using second-order all pole modelling, and a technique for extracting FM features using this model is proposed, to produce very smooth, slowly varying FM features that are effective for speaker recognition. This is evaluated on NIST 2008 international speaker recognition evaluation. Optimum Frequency band allocation specifically to capture speaker specific information is studied in terms of the number of sub-bands and spacing of center frequencies, and two new frequency band re-allocations are proposed for FM based speaker recognition. Furthermore, FM feature is successfully utilized for automatic forensic speaker recognition, which is implemented based on the likelihood ratio framework with two stage modelling and calibration, and shown to behave in a complementary manner to amplitude information.
Vairāk informācijas
| Autors | Julien Epps, Tharmarajah Thiruvaran, Eliathamby Ambikairajah |
|---|---|
| Izdevējs | Scholars' Press |
| Izlaides gads | 2014 |
| Vāka tips | Mīkstais vāks |
| EAN | 9783639708141 |