1 |
What's "up"? Impaired Spatial Preposition Processing in Posterior Cortical Atrophy.
|
|
|
|
BASE
|
|
Show details
|
|
4 |
What's "up"? Impaired Spatial Preposition Processing in Posterior Cortical Atrophy. ...
|
|
|
|
BASE
|
|
Show details
|
|
5 |
Auditory distraction while reading in different languages ...
|
|
|
|
BASE
|
|
Show details
|
|
6 |
Topic-Oriented Text Features Can Match Visual Deep Models of Video Memorability
|
|
|
|
In: Applied Sciences ; Volume 11 ; Issue 16 (2021)
|
|
BASE
|
|
Show details
|
|
7 |
Structural acoustic features of human musicality scored on primate vocalizations ...
|
|
|
|
BASE
|
|
Show details
|
|
8 |
A new Unenlagiinae (Theropoda: Dromaeosauridae) from the Late Cretaceous of Brazil ...
|
|
|
|
BASE
|
|
Show details
|
|
9 |
Use of the electronic tongue as a tool for the characterization of Melipona scutellaris Latreille honey
|
|
|
|
BASE
|
|
Show details
|
|
10 |
Comparison of Word2vec with Hash2vec for Machine Translation
|
|
|
|
In: Master's Projects (2020)
|
|
BASE
|
|
Show details
|
|
11 |
The relevance of valence orientation for clustering three Niger-Congo language families
|
|
|
|
In: https://hal.archives-ouvertes.fr/hal-02429737 ; 2019 (2019)
|
|
BASE
|
|
Show details
|
|
12 |
The relevance of valence orientation for clustering three Niger-Congo language families
|
|
|
|
In: https://hal.archives-ouvertes.fr/hal-02429737 ; 2019 (2019)
|
|
BASE
|
|
Show details
|
|
13 |
Image Processing Based Speaking System For Mute People Using Hand Gestures ...
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Image Processing Based Speaking System For Mute People Using Hand Gestures ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
A visual familiarity account of evidence for orthographic processing in pigeons (Columbia livia): a reply to Scarf, Corballis, Güntürkün, and Colombo (2017)
|
|
|
|
BASE
|
|
Show details
|
|
16 |
Estudio de la influencia de incorporar conocimiento léxico-semántico a la técnica de Análisis de Componentes Principales para la generación de resúmenes multilingües
|
|
|
|
In: Linguamática, Vol 7, Iss 1 (2015) (2015)
|
|
BASE
|
|
Show details
|
|
17 |
Parallel ICA of FDG-PET and PiB-PET in three conditions with underlying Alzheimer's pathology.
|
|
|
|
BASE
|
|
Show details
|
|
18 |
Parallel ICA of FDG-PET and PiB-PET in three conditions with underlying Alzheimer's pathology.
|
|
|
|
BASE
|
|
Show details
|
|
19 |
Discourse and Prosody in spoken French: Why, what and how should one count? A comparative statistical perspective
|
|
|
|
In: ISSN: 1661-3171 ; EISSN: 0259-6199 ; Cahiers de linguistique française ; SWIP 3 - Swiss Workshop In Prosody Proceedings - Actes des journées Genève septembre 2014 ; SWIP 3 - Swiss Workshop In Prosody ; https://halshs.archives-ouvertes.fr/halshs-01066796 ; SWIP 3 - Swiss Workshop In Prosody, Sep 2014, Switzerland. pp.33-44 (2014)
|
|
BASE
|
|
Show details
|
|
20 |
Combination of Cepstral and Phonetically Discriminative Features for Speaker Verification
|
|
|
|
In: ISSN: 1070-9908 ; IEEE Signal Processing Letters ; https://hal.archives-ouvertes.fr/hal-01690336 ; IEEE Signal Processing Letters, Institute of Electrical and Electronics Engineers, 2014, 21 (9), pp.1040 - 1044. ⟨10.1109/LSP.2014.2323432⟩ (2014)
|
|
Abstract:
International audience ; Most speaker recognition systems rely on short-term acoustic cepstral features for extracting the speaker-relevant information from the signal. But phonetic discriminant features, extracted by a bottleneck multi-layer perceptron (MLP) on longer stretches of time, can provide a complementary information and have been adopted in speech transcription systems. We compare the speaker verification performance using cepstral features, discriminant features, and a concatenation of both followed by a dimension reduction. We consider two speaker recognition systems, one based on maximum likelihood linear regression (MLLR) super-vectors and the other on a state-of-the-art i-vector system with two session variability compensation schemes. Experiments are reported on a standard configuration of NIST SRE 2008 and 2010 databases. The results show that the phonetically discriminative MLP features retain speaker-specific information which is complementary to the short-term cepstral features. The performance improvement is obtained with both score domain and feature domain fusion and the speaker verification equal error rate (EER) is reduced up to 50% relative, compared to the best i-vector system using only cepstral features.
|
|
Keyword:
[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing; [INFO]Computer Science [cs]; bottleneck features; i-vector; LDA; multi-layer perceptron; PCA; PLDA; Speaker verification
|
|
URL: https://doi.org/10.1109/LSP.2014.2323432 https://hal.archives-ouvertes.fr/hal-01690336/document https://hal.archives-ouvertes.fr/hal-01690336/file/double-final.pdf https://hal.archives-ouvertes.fr/hal-01690336
|
|
BASE
|
|
Hide details
|
|
|
|