150,395 research outputs found
Speaker segmentation and clustering
No full text07.08.13 KB. Ok to add the accepted version to Spiral, Elsevier says ok whlile mandate not enforced.This survey focuses on two challenging speech processing topics, namely: speaker segmentation and speaker clustering. Speaker segmentation aims at finding speaker change points in an audio stream, whereas speaker clustering aims at grouping speech segments based on speaker characteristics. Model-based, metric-based, and hybrid speaker segmentation algorithms are reviewed. Concerning speaker clustering, deterministic and probabilistic algorithms are examined. A comparative assessment of the reviewed algorithms is undertaken, the algorithm advantages and disadvantages are indicated, insight to the algorithms is offered, and deductions as well as recommendations are given. Rich transcription and movie analysis are candidate applications that benefit from combined speaker segmentation and clustering. © 2007 Elsevier B.V. All rights reserved
Recommended from our members
Biologically inspired speaker verification
No full textSpeaker verification is an active research problem that has been addressed using a variety of different classification techniques. However, in general, methods inspired by the human auditory system tend to show better verification performance than other methods. In this thesis three biologically inspired speaker verification algorithms are presented
Speaker Diarization: Its Developments, Applications, And Challenges
No full textMultimedia computing is an emerging research area with a wide range of possibilities for various types of digital media. For instance television, movie, audio broadcast, and meeting recordings can be analysed to derive semantic value. Although significant progress has been achieved in recent years, speaker diarization continues to be an active topic in speech research. In this paper, a speaker diarization system would be reviewed and then continued by description of current development and application of diarization system in the broadcast news, meeting and telephone channel. Some trends and challenges arising in the speaker diarization area are addressed in this review, such as presence of overlapped speech and various characteristic of the input audio stream. Particularly, performance improvement has become an ongoing goal in speaker diarization research. One of important factors is how far speaker diarization system has been implemented in many widely area to assist other application systems.
Keywords : component; speaker diarization, broadcast news, meeting, telephone channel, overlapped speec
Speaker adaptation and the evaluation of speaker similarity in the EMIME speech-to-speech translation project
No full textThis paper provides an overview of speaker adaptation research carried out in the EMIME speech-to-speech translation (S2ST) project. We focus on how speaker adaptation transforms can be learned from speech in one language and applied to the acoustic models of another language. The adaptation is transferred across languages and/or from recognition models to synthesis models. The various approaches investigated can all be viewed as a process in which a mapping is defined in terms of either acoustic model states or linguistic units. The mapping is used to transfer either speech data or adaptation transforms between the two models. Because the success of speaker adaptation in text-to-speech synthesis is measured by judging speaker similarity, we also discuss issues concerning evaluation of speaker similarity in an S2ST scenario
Cross-Lingual Speaker Discrimination Using Natural and Synthetic Speech
No full textThis paper describes speaker discrimination experiments in which native English listeners were presented with either natural speech stimuli in English and Mandarin, synthetic speech stimuli in English and Mandarin, or natural Mandarin speech and synthetic English speech stimuli. In each experiment, listeners were asked to decide whether they thought the sentences were spoken by the same person or not. We found that the results for Mandarin/English speaker discrimination are very similar to results found in previous work on German/English and Finnish/English speaker discrimination. We conclude from this and previous work that listeners are able to identify speakers across languages and they are able to identify speakers across speech types, but the combination of these two factors leads to a speaker discrimination task which is too difficult for listeners to perform successfully, given the quality of across-language speaker adapted speech synthesis at present
Speaker similarity evaluation of foreign-accented speech synthesis using HMM-based speaker adaptation
No full textThis paper describes a speaker discrimination experiment in which native English listeners were presented with natural and synthetic speech stimuli in English and were asked to judge whether they thought the sentences were spoken by the same person or not. The natural speech consisted of recordings of Finnish speakers speaking English. The synthetic stimuli were created using adaptation data from the same Finnish speakers. Two average voice models were compared: one trained on Finnish-accented English and the other on American-accented English. The experiments illustrate that listeners perform well at speaker discrimination when the stimuli are both natural or both synthetic, but when the speech types are crossed performance drops significantly. We also found that the type of accent in the average voice model had no effect on the listeners' speaker discrimination performance
Efficient Invariant Features for Sensor Variability Compensation in Speaker Recognition
No full textIn this paper, we investigate the use of invariant features for speaker recognition. Owing to their characteristics, these features are introduced to cope with the difficult and challenging problem of sensor variability and the source of performance degradation inherent in speaker recognition systems. Our experiments show: (1) the effectiveness of these features in match cases; (2) the benefit of combining these features with the mel frequency cepstral coefficients to exploit their discrimination power under uncontrolled conditions (mismatch cases). Consequently, the proposed invariant features result in a performance improvement as demonstrated by a reduction in the equal error rate and the minimum decision cost function compared to the GMM-UBM speaker recognition systems based on MFCC features
Evaluation of the Vulnerability of Speaker Verification to Synthetic Speech
No full textIn this paper, we evaluate the vulnerability of a speaker verification
(SV) system to synthetic speech. Although this problem
was first examined over a decade ago, dramatic improvements
in both SV and speech synthesis have renewed interest in
this problem. We use a HMM-based speech synthesizer, which
creates synthetic speech for a targeted speaker through adaptation
of a background model and a GMM-UBM-based SV system.
Using 283 speakers from the Wall-Street Journal (WSJ)
corpus, our SV system has a 0.4% EER. When the system
is tested with synthetic speech generated from speaker models
derived from the WSJ journal corpus, 90% of the matched
claims are accepted. This result suggests a possible vulnerability
in SV systems to synthetic speech. In order to detect
synthetic speech prior to recognition, we investigate the
use of an automatic speech recognizer (ASR), dynamic-timewarping
(DTW) distance of mel-frequency cepstral coefficients
(MFCC), and previously-proposed average inter-frame difference
of log-likelihood (IFDLL). Overall, while SV systems
have impressive accuracy, even with the proposed detector,
high-quality synthetic speech can lead to an unacceptably high
acceptance rate of synthetic speakers
New consonantal acoustic parameters for forensic speaker comparison
No full textThis thesis examines acoustic parameters of five consonants /m, n, ŋ, l, s/ in two dialects of British English: Standard Southern British English and Leeds English. The research aims to explore population distributions of the acoustic features, gauge cross-dialectal variation, and discover new parameters for application in forensic speaker comparison casework. The five parameters investigated for each segment are:
For /m, n, ŋ, l/:
Normalised duration, Centre of gravity, Standard deviation, Frequency at peak amplitude, Frequency at minimum amplitude
For /s/: Normalised duration, Centre of gravity, Standard deviation, Skewness, Kurtosis
The work contributes firstly to the general phonetic literature by presenting acoustic data for a number of parameters and consonant segments that have not been previously studied in depth in these dialects. Secondly, the research informs the forensic phonetic literature by considering the intra- and inter-speaker variability and gauging the relative speaker-specificity of each acoustic feature. Discriminant analysis and likelihood ratio estimation assess the discrimination ability of each feature, and results highlight several promising parameters with potential for application in forensic speaker comparison casework
Speaker Diarization Based on Intensity Channel Contribution
No full textThe time delay of arrival (TDOA) between multiple microphones has been used since 2006 as a source of information (localization) to complement the spectral features for speaker diarization. In this paper, we propose a new localization feature, the intensity channel contribution (ICC) based on the relative energy of the signal arriving at each channel compared to the sum of the energy of all the channels. We have demonstrated that by joining the ICC features and the TDOA features, the robustness of the localization features is improved and that the diarization error rate (DER) of the complete system (using localization and spectral features) has been reduced. By using this new localization feature, we have been able to achieve a 5.2% DER relative improvement in our development data, a 3.6% DER relative improvement in the RT07 evaluation data and a 7.9% DER relative improvement in the last year's RT09 evaluation data
- …
