Teva Merlin

A tutorial on text-independent speaker verification

This paper presents an overview of a state-of-the-art text-independent speaker verification system. First, an introduction proposes a modular scheme of the training and test phases of a speaker verification system. Then, the most commonly speech parameterization used in speaker verification, namely, cepstral analysis, is detailed. Gaussian mixture modeling, which is the speaker modeling technique used in most systems, is then explained. A few speaker modeling alternatives, namely, neural networks and support vector machines, are mentioned. Normalization of scores is then explained, as this is a very important step to deal with real-world data. The evaluation of a speaker verification system is then detailed, and the detection error trade-off (DET) curve is explained. Several extensions of speaker verification are then enumerated, including speaker tracking and segmentation by speakers. Then, some applications of speaker verification are proposed, including on-site applications, remote applications, applications relative to structuring audio information, and games. Issues concerning the forensic area are then recalled, as we believe it is very important to inform people about the actual performance and limitations of speaker verification systems. This paper concludes by giving a few research trends in speaker verification for the next couple of years.

A speaker tracking system based on speaker turn detection for NIST evaluation

A speaker tracking system (STS) is built by using successively a speaker change detector and a speaker verification system. The aim of the STS is to find in a conversation between several persons (some of them having already enrolled and other being totally unknown) target speakers chosen in a set of enrolled users. In a first step, speech is segmented into homogeneous segments containing only one speaker, without any use of a priori knowledge about speakers. Then, the resulting segments are checked to belong to one of the target speakers. The system has been used in a NIST evaluation test with satisfactory results.

Evolutive HMM for multi-speaker tracking system

Seeking within a speech sequence the speaker utterances is one of the main tasks of indexing. In this paper, the proposed speaker tracking system is defined in the case where all speaker identities are known beforehand. The conversation is modeled as an evolutive HMM-like model, in which speaker models computed are added one by one. A temporary indexing is proposed after each speaker adding and then challenged at the next step. This process is iterated until all the speakers are detected. The system has been assessed using multi-speaker messages generated by concatenation of Switchboard mono-speaker segments. The obtained results show the potentiality of the proposed solution.

AMIRAL: A Block-Segmental Multirecognizer Architecture for Automatic Speaker Recognition

Abstract Fredouille, Corinne, Bonastre, Jean-Francois, and Merlin, Teva, AMIRAL: A Block-Segmental Multirecognizer Architecture for Automatic Speaker Recognition, Digital Signal Processing10(2000), 172–197. In the wide domain of automatic speech recognition, extracting the relevant information carried by the speech signal is far from easy. Diversity, redundancy, and variability, the main characteristics of the speech signal, make this task particularly difficult. The work reported here presents a multirecognizer architecture designed to cope with this issue in the framework of Automatic Speaker Recognition. This architecture, based on various individual recognizers, exploits different classes of information conveyed by the speech signal. In this paper, two classes of information are investigated: information related to the frequency domain, and “dynamic” information. This multirecognizer architecture is coupled with a block-segmental approach applied on each classifier. The overall system allows us to emphasize the most informative temporal blocks and to discard the least informative ones or those corrupted by noise. The AMIRAL system developed by the LIA integrates both approaches and was tested during the NIST/NSA 1999 speaker recognition evaluations. The results of these experiments for the tasks of Speaker Verification (“One Speaker” and “Two Speakers”) and Speaker Tracking are provided and discussed.

Computer-assisted transcription of speech based on confusion network reordering

Large vocabulary automatic speech recognition (ASR) technologies perform well in known and controlled contexts. In less controlled conditions, however, human review is often necessary to check and correct the results of such systems in order to ensure that the output of ASR will be understandable. We propose a method for computer-assisted transcription of speech, based on automatic reordering confusion networks. Our method will be evaluated in terms of KSR (Keystroke Saving Rate) and WSR (Word Stroke Ratio). It allows to significantly reduce the number of actions needed to correct ASR outputs. WSR computed before and after every network reordering shows a gain of about 17.7% (3.4 points).

Iterative filtering of phonetic transcriptions of proper nouns

This paper focuses on an approach to enhancing automatic phonetic transcription of proper nouns by using an iterative filter to retain only the most relevant part of a large set of phonetic variants, obtained by combining rule-based generation with extraction from actual audio signals. Using this technique, we were able to reduce the error rate affecting proper nouns during automatic speech transcription of the ESTER corpus of French broadcast news. The role of the filtering was to ensure that the new phonetic variants of proper nouns would not induce new errors in the transcription of the rest of the words.

Speaker detection using multi-speaker audio files for both enrollment and test

This paper focuses on speaker detection using multispeaker files both for the enrollment phase and for the test phase. This task was introduced during the 2002 NIST speaker recognition evaluation campaign. Enrollment data is composed of three two-speaker files. Test files are also two-speaker records. The system presented here uses a speaker segmentation process based on an HMM conversation model followed by a speaker matching technique to produce one-speaker segments. Speaker detection is then achieved using AMIRAL, LIAs GMM-based speaker verification system. Validation of the proposed strategy is done using extracts from the NIST 2002 results.