Speaker-adaptive acoustic-articulatory inversion using cascaded gaussian mixture regression
This paper addresses the adaptation of an acoustic-articulatory model of a reference
speaker to the voice of another speaker, using a limited amount of audio-only data. In the …
speaker to the voice of another speaker, using a limited amount of audio-only data. In the …
Semi-automatic video annotation based on active learning with multiple complementary predictors
In this paper, we will propose a novel semi-automatic annotation scheme for video semantic
classification. It is well known that the large gap between high-level semantics and low-level …
classification. It is well known that the large gap between high-level semantics and low-level …
[PDF][PDF] Spoken Document Retrieval for TREC-9 at Cambridge University.
SE Johnson, P Jourlin, KS Jones, PC Woodland - TREC, 2000 - trec.nist.gov
This paper presents work done at Cambridge University for the TREC-9 Spoken Document
Retrieval (SDR) track. The CUHTK transcriptions from TREC-8 with Word Error Rate (WER) …
Retrieval (SDR) track. The CUHTK transcriptions from TREC-8 with Word Error Rate (WER) …
22 Question Answering
B Webber, N Webb - The handbook of computational linguistics …, 2010 - Wiley Online Library
Questions are asked and answered every day. Question answering (QA) technology aims to
deliver the same facility online. It goes further than the more familiar search based on …
deliver the same facility online. It goes further than the more familiar search based on …
Method and system for cross-lingual voice conversion
I Agiomyrgiannakis - US Patent 9,177,549, 2015 - Google Patents
US 2015/O127349 A1 May 7, 2015 (57) ABSTRACT (51) Int. Cl. A method and system for is
disclosed for cross-lingual Voice GOL 5/00(2013.01) conversion. A speech-to-speech …
disclosed for cross-lingual Voice GOL 5/00(2013.01) conversion. A speech-to-speech …
Comparison of speaker adaptation methods as feature extraction for SVM-based speaker recognition
In the last years the speaker recognition field has made extensive use of speaker adaptation
techniques. Adaptation allows speaker model parameters to be estimated using less speech …
techniques. Adaptation allows speaker model parameters to be estimated using less speech …
A basis representation of constrained MLLR transforms for robust adaptation
Abstract Constrained Maximum Likelihood Linear Regression (CMLLR) is a speaker
adaptation method for speech recognition that can be realized as a feature-space …
adaptation method for speech recognition that can be realized as a feature-space …
Lightly supervised and data-driven approaches to mandarin broadcast news transcription
B Chen, JW Kuo, WH Tsai - 2004 IEEE International …, 2004 - ieeexplore.ieee.org
This paper investigates the use of several lightly supervised and data-driven approaches to
Mandarin broadcast news transcription. First, with a consideration of the special structural …
Mandarin broadcast news transcription. First, with a consideration of the special structural …
A novel channel estimate for noise robust speech recognition
G Vanderreydt, K Demuynck - Computer Speech & Language, 2024 - Elsevier
We propose a novel technique to estimate the channel characteristics for robust speech
recognition. The method focuses on reliable time–frequency speech patches which are …
recognition. The method focuses on reliable time–frequency speech patches which are …
Frequency warping for VTLN and speaker adaptation by linear transformation of standard MFCC
S Panchapagesan, A Alwan - Computer speech & language, 2009 - Elsevier
Vocal tract length normalization (VTLN) for standard filterbank-based Mel frequency cepstral
coefficient (MFCC) features is usually implemented by warping the center frequencies of the …
coefficient (MFCC) features is usually implemented by warping the center frequencies of the …