Automatic Language Identification Research Articles

Language Identification: A Tutorial

This tutorial presents an overview of the progression of spoken language identification (LID) systems and current developments. The introduction provides a background on automatic language identification systems using syntactic, morphological, and in particular, acoustic, phonetic, phonotactic and prosodic level information. Different frontend features that are used in LID systems are presented. Several normalization and language modelling techniques have also been presented. We also discuss different LID system architectures that embrace a variety of front-ends and back-ends, and configurations such as hierarchical and fusion classifiers. Evaluations of the LID system are presented using NIST language recognition evaluation tasks.

A Flexible Classifier Design Framework Based on Multiobjective Programming

We propose a multiobjective programming (MOP) framework for finding compromise solutions that are satisfactory for each of multiple competing performance criteria in a pattern classification task. The fundamental idea for our formulation of classifier learning, which we refer to as iterative constrained optimization (ICO), evolves around improving one objective while allowing the rest to degrade. This is achieved by the optimization of individual objectives with proper constraints on the remaining competing objectives. The constraint bounds are adjusted based on the objective functions obtained in the most recent iteration. An aggregated utility function is used to evaluate the acceptability of local changes in competing criteria, i.e., changes from one iteration to the next. Although many MOP approaches developed so far are formal and extensible to large number of competing objectives, their capabilities are examined only with two or three objectives. This is mainly because practical problems become significantly harder to manage when the number of objectives gets larger. We, however, illustrate the proposed framework in the context of automatic language identification (LID) of 12 languages and three dialects. This LID task requires the simultaneous minimization of the false-acceptance and false-rejection rates for each of the 15 languages/dialects, and, hence, is an MOP problem with a total of 30 competing objectives. In our experiments, we observed that the ICO-trained classifiers result in not only reduced error rates but also a good balance among the many competing objectives when compared to those classifiers that minimize an overall objective. We interpret our experimental findings as evidence for ICO offering a greater degree of freedom for classifier design.

Channel mitigation approach for automatic language identification

A major obstacle to overcome in language identification (LID) performance is the impact of varying channel conditions. In house experiments show that LID performance drops between 10‐12% across channels. The focus of this project is to mitigate the impact of channel conditions and artifacts on LID, and to provide an understanding of how channel robust models may be created by combining data from across corpora. Our approach involved creating composite cross‐channel language models from multiple corpora that were tested with data from three corpora whose results were compared to results obtained from same‐channel and pure cross‐channel experiments. Our hypotheses were that 1) same‐channel models would be the most accurate, 2) purely cross‐channel models would be considerably less accurate, and 3) composite model accuracy would fall in between that of the same‐channel and cross‐channel models. Results were surprising: while pure cross‐channel tests performed the worst, with an average of 11% loss in accuracy, composite models outperformed matched condition models in tests by 0.5% to 15% for all four combinations tested. These findings provide evidence that cross‐channel augmentation does not just make for more robust LID models across channel conditions, but improves model robustness within channel conditions as well.

Automatic Language Identification with Discriminative Language Characterization Based on SVM

Robust automatic language identification (LID) is the task of identifying the language from a short utterance spoken by an unknown speaker. The mainstream approaches include parallel phone recognition language modeling (PPRLM), support vector machine (SVM) and the general Gaussian mixture models (GMMs). These systems map the cepstral features of spoken utterances into high level scores by classifiers. In this paper, in order to increase the dimension of the score vector and alleviate the inter-speaker variability within the same language, multiple data groups based on supervised speaker clustering are employed to generate the discriminative language characterization score vectors (DLCSV). The back-end SVM classifiers are used to model the probability distribution of each target language in the DLCSV space. Finally, the output scores of back-end classifiers are calibrated by a pair-wise posterior probability estimation (PPPE) algorithm. The proposed language identification frameworks are evaluated on 2003 NIST Language Recognition Evaluation (LRE) databases and the experiments show that the system described in this paper produces comparable results to the existing systems. Especially, the SVM framework achieves an equal error rate (EER) of 4.0% in the 30-second task and outperforms the state-of-art systems by more than 30% relative error reduction. Besides, the performances of proposed PPRLM and GMMs algorithms achieve an EER of 5.1% and 5.0% respectively.

Using SVM as Back-End Classifier for Language Identification

Robust automatic language identification (LID) is a task of identifying the language froma short utterance spoken by an unknown speaker. One of the mainstream approaches named parallel phone recognition language modeling (PPRLM) has achieved a very good performance. The log-likelihood radio (LLR) algorithm has been proposed recently to normalize posteriori probabilities which are the outputs of back-end classifiers in PPRLM systems. Support vector machine (SVM) with radial basis function (RBF) kernel is adopted as the back-end classifier. But for the conventional SVM classifier, the output is not probability. We use a pairwise posterior probability estimation (PPPE) algorithm to calibrate the output of each classifier. The proposed approaches are evaluated on the 2005 National Institute of Standards and Technology (NIST). Language recognition evaluation databases and experiments show that the systems described in this paper produce comparable results to the existing arts.

Automatic Prosodic Variations Modeling for Language and Dialect Discrimination

This paper addresses the problem of modeling prosody for language identification. The aim is to create a system that can be used prior to any linguistic work to show if prosodic differences among languages or dialects can be automatically determined. In previous papers, we defined a prosodic unit, the pseudosyllable. Rhythmic modeling has proven the relevance of the pseudosyllable unit for automatic language identification. In this paper, we propose to model the prosodic variations, that is to say model sequences of prosodic units. This is achieved by the separation of phrase and accentual components of intonation. We propose an independent coding of those components on differentiated scales of duration. Short-term and long-term language-dependent sequences of labels are modeled by n-gram models. The performance of the system is demonstrated by experiments on read speech and evaluated by experiments on spontaneous speech. Finally, an experiment is described on the discrimination of Arabic dialects, for which there is a lack of linguistic studies, notably on prosodic comparisons. We show that our system is able to clearly identify the dialectal areas, leading to the hypothesis that those dialects have prosodic differences.

A Vector Space Modeling Approach to Spoken Language Identification

We propose a novel approach to automatic spoken language identification (LID) based on vector space modeling (VSM). It is assumed that the overall sound characteristics of all spoken languages can be covered by a universal collection of acoustic units, which can be characterized by the acoustic segment models (ASMs). A spoken utterance is then decoded into a sequence of ASM units. The ASM framework furthers the idea of language-independent phone models for LID by introducing an unsupervised learning procedure to circumvent the need for phonetic transcription. Analogous to representing a text document as a term vector, we convert a spoken utterance into a feature vector with its attributes representing the co-occurrence statistics of the acoustic units. As such, we can build a vector space classifier for LID. The proposed VSM approach leads to a discriminative classifier backend, which is demonstrated to give superior performance over likelihood-based n-gram language modeling (LM) backend for long utterances. We evaluated the proposed VSM framework on 1996 and 2003 NIST Language Recognition Evaluation (LRE) databases, achieving an equal error rate (EER) of 2.75% and 4.02% in the 1996 and 2003 LRE 30-s tasks, respectively, which represents one of the best results reported on these popular tasks

Language identification using acoustic log-likelihoods of syllable-like units

Automatic spoken language identification (LID) is the task of identifying the language from a short utterance of the speech signal uttered by an unknown speaker. The most successful approach to LID uses phone recognizers of several languages in parallel [Zissman, M.A., 1996. Comparison of four approaches to automatic language identification of telephone speech. IEEE Trans. Speech Audio Process. 4 (1), 31–44]. The basic requirement to build a parallel phone recognition (PPR) system is segmented and labeled speech corpora. In this paper, a novel approach is proposed for the LID task which uses parallel syllable-like unit recognizers, in a frame work similar to the PPR approach in the literature. The difference is that the sub-word unit models for each of the languages to be recognized are generated in an unsupervised manner without the use of segmented and labeled speech corpora. The training data of each of the languages is first segmented into syllable-like units and language-dependent syllable-like unit inventory is created. These syllable-like units are then clustered using an incremental approach. This results in a set of syllable-like units models for each language. Using these language-dependent syllable-like unit models, language identification is performed based on accumulated acoustic log-likelihoods. Our initial results on the Oregon Graduate Institute Multi-language Telephone Speech Corpus [Muthusamy, Y.K., Cole, R.A., Oshika, B.T., 1992. The OGI multi-language telephone speech corpus. In: Proceedings of Internat. Conf. Spoken Language Process., October 1992, pp. 895–898] show that the performance is 72.3%. We further show that if only a subset of syllable-like unit models that are unique (in some sense) are considered, the performance improves to 75.9%.

Rhythmic unit extraction and modelling for automatic language identification

This paper deals with an approach to automatic language identification based on rhythmic modelling. Beside phonetics and phonotactics, rhythm is actually one of the most promising features to be considered for language identification, even if its extraction and modelling are not a straightforward issue. Actually, one of the main problems to address is what to model. In this paper, an algorithm of rhythm extraction is described: using a vowel detection algorithm, rhythmic units related to syllables are segmented. Several parameters are extracted (consonantal and vowel duration, cluster complexity) and modelled with a Gaussian Mixture. Experiments are performed on read speech for seven languages (English, French, German, Italian, Japanese, Mandarin and Spanish) and results reach up to 86 ± 6% of correct discrimination between stress-timed mora-timed and syllable-timed classes of languages, and to 67 ± 8% of correct language identification on average for the seven languages with utterances of 21 s. These results are commented and compared with those obtained with a standard acoustic Gaussian mixture modelling approach (88 ± 5% of correct identification for the seven languages identification task).

Unified messaging system with automatic language identification for text-to-speech conversion

A unified messaging system includes a voice gateway server coupled to an electronics mail system and a private branch exchange (PBX). The voice gateway server provides voice messaging services to a set of subscribers. Within the voice gateway server, a trigraph analyzer sequentially examines 3-character combinations within a text message; determines occurrence frequencies for the character combinations; compares the occurrence frequencies with reference occurrence statistics modeled from text samples written in particular languages; and generates a language identifier and a likelihood value for the text message. Based upon the language identifier, a message inquiry unit selects an appropriate text-to-speech engine for converting the text message into computer-generated speech that is played to a subscriber.

Automatic language identification

Automatic language identification of speech is the process by which the language of a digitized speech utterance is recognized by a computer. In this paper, we will describe the set of available cues for language identification of speech and discuss the different approaches to building working systems. This overview includes a range of historical approaches, contemporary systems that have been evaluated on standard databases, and promising future approaches. Comparative results are also reported.

The use of cognitive processing strategies and linguistic cues for efficient automatic language identification

Automatic language identification is clearly an area of research which requires much interdisciplinary effort. However, the major part of the research is carried out by the speech recognition/signal processing community where automatic language identification is seen as being related to speaker independent speech recognition and speaker identification. In particular, speaker identification methods appear to outperform all other methods and the incorporation of prosodic information has contributed only marginally to their success. This is a counterintuitive result that ignores the wealth of data made available by both the linguistic and psycholinguistic communities. The current work presents a methodology based on the cognitive processing strategies which human beings use. In particular, it is proposed that correctly parameterized, prosodic information, such as might be associated with a language with which the listener has little familiarity, is an important discriminating feature. In addition, the hierarchical approach to discrimination which people use, together with suitably parameterized linguistic cues would be more computationally efficient than the indiscriminate extraction and analysis of statistical parameters. In order to illustrate the approach a data set was chosen such as to incorporate examples of what appear to native English speakers to be two quite similar languages and one that is very different from both: Spanish, Portuguese and Chinese. The Hurst exponent is used as a novel parameterization of the prosodic signature resulting from intonation effects over a phrase or short sentence. The results of the illustrative study are in keeping with our own abilities to discriminate between those languages. They indicate that the perception of global, prosodic information subsumes the effect of similarities which may exist between the languages at the local level of acoustic features.

Automatic language identification: an alternative approach to phonetic modelling

This paper deals with our research on vowel system modelling in an automatic language identification (LID) purpose. The study of vowel systems shows that they carry an important part of the language characteristics. Taking advantage of this knowledge is promising. We propose an alternative modelling to the standard acoustic–phonetic decoding currently used as front-end in the LID systems: each language vocalic system is modelled by a Gaussian mixture, estimated out from automatically detected vowels. OGI multi-lingual telephone speech (MLTS) corpus is used to assess this approach. In a five language close set identification task, we reach 57.3% of correct identification with 45 s duration utterances. Taking into account that we use only the vowel information (on average in our database, 45 s of speech consist of less than 25 s of vowels, 10 s of consonants and 10–15 s of silence) and no language modelling, these results are very promising and offer many perspectives.

Language identification with suprasegmental cues: a study based on speech resynthesis.

This paper proposes a new experimental paradigm to explore the discriminability of languages, a question which is crucial to the child born in a bilingual environment. This paradigm employs the speech resynthesis technique, enabling the experimenter to preserve or degrade acoustic cues such as phonotactics, syllabic rhythm, or intonation from natural utterances. English and Japanese sentences were resynthesized, preserving broad phonotactics, rhythm, and intonation (condition 1), rhythm and intonation (condition 2), intonation only (condition 3), or rhythm only (condition 4). The findings support the notion that syllabic rhythm is a necessary and sufficient cue for French adult subjects to discriminate English from Japanese sentences. The results are consistent with previous research using low-pass filtered speech, as well as with phonological theories predicting rhythmic differences between languages. Thus, the new methodology proposed appears to be well suited to study language discrimination. Applications for other domains of psycholinguistic research and for automatic language identification are considered.

Segment-based automatic language identification

This paper discusses the formulation, development and analysis of a segment-based approach to the automatic language identification (LID) problem. This system utilizes phonotactic, acoustic-phonetic, and prosodic information within a unified probabilistic framework. The implementation of this framework allows the relative contributions of different sources of information to be determined empirically, as well as providing the mechanism for combining them within one system. The system has been evaluated using the Oregon Graduate Institute (OGI) multi-language telephone speech corpus and the results are competitive with other current LID systems. The results have also indicated that, while the phontotactic information of a spoken utterance is the most useful information for LID, acoustic-phonetic and prosodic information can be useful for increasing a system’s accuracy, especially when the utterance is short.

Determination of the script and language content of document images

Most document recognition work to date has been performed on English text. Because of the large overlap of the character sets found in English and major Western European languages such as French and German, some extensions of the basic English capability to those languages have taken place. However, automatic language identification prior to optical character recognition is not commonly available and adds utility to such systems. Languages and their scripts have attributes that make it possible to determine the language of a document automatically. Detection of the values of these attributes requires the recognition of particular features of the document image and, in the case of languages using Latin-based symbols, the character syntax of the underlying language. We have developed techniques for distinguishing which language is represented in an image of text. This work is restricted to a small but important subset of the world's languages. The method first classifies the script into two broad classes: Han-based and Latin-based. This classification is based on the spatial relationships of features related to the upward concavities in character structures. Language identification within the Han script class (Chinese, Japanese, Korean) is performed by analysis of the distribution of optical density in the text images. We handle 23 Latin-based languages using a technique based on character shape codes, a representation of Latin text that is inexpensive to compute.

Language and dialect identification by syllabic spectral features

A speaker-based automatic language identification system has been studied to discriminate both languages and dialects. Syllabic spectral features are extracted automatically by an artificial neural network. The system makes no use of phonetic segmentation for recognition or training. High performance has been obtained for monologue telephone recordings. The system has been applied to several different databases to study the discrimination possible among different languages, dialects, and accents with such limited phonetic–acoustic information. The system performance is also related to individual speaker characteristics and the size of reference population for each language, and the length of testing samples. This study will report experimental results related to performance and data environments. It will also report an attempt by others to compare recognition performance of trained or expert human listeners with automatic recognition results.

Development of an approach to automatic language identification based on phone recognition

An Automatic Language Identification (LID) approach is presented. The baseline LID system consists of three parts: (1) hidden Markov model (HMM) based context-independent phone recognizers, (2) language identification score generators and (3) a linear language classifier. The system exploits language-dependent phonotactic constraints and prosodic information. Four methods are proposed to improve the system performance. Two bigram-based interpolated N-gram language models (forward and backward) are used to model the phone sequence constraints of different spoken languages. A context-dependent duration model interpolated by a context-independent duration model is used to capture the duration information. Comparison experiments between the linear classifier and neural network-based final classifiers were conducted. Finally, optimization of language model based on back propagation is proposed. The improved system was evaluated on an 11-language task, and performance reached 13·3% and 26·2% (error rate) for utterances averaging 45 s duration and 10 s duration, respectively. Compared with the baseline system performance, it shows the importance of the issues addressed in this paper for language identification.

Comparison of four approaches to automatic language identification of telephone speech

Abstruct- We have compared the performance of four approaches for automatic language identification of speech utterances: Gaussian mixture model (GMM) classification; single-language phone recognition followed by languagedependent, interpolated n-gram language modeling (PRLM); parallel PRLM, which uses multiple single-language phone recognizers, each trained in a different language; and languagedependent parallel phone recognition (PPR). These approaches, which span a wide range of training requirements and levels of recognition complexity, were evaluated with the Oregon Graduate Institute Multi-Language Telephone Speech Corpus. Systems containing phone recognizers performed better than the simpler GMM classifier. The top-performing system was parallel PRLM, which exhibited an error rate of 2% for 45-s utterances and 5% for 10-s utterances in two-language, closed-set, forcedchoice classification. The error rate for 11-language, closed-set, forced-choice classification was 11 % for 45-s utterances and 21% for 10-s utterances.

An investigation of approaches to language identification

This paper will present a discussion of approaches used in the area of language identification. Past efforts have attempted to develop an automatic language identification system based upon the employment of linguistic units, i.e., phones or phonemes, with moderate results. An approach currently under investigation at SCRL is the application of pattern analysis techniques to features extracted from the speech signal, without regard to specific linguistic units. Upon completion of the first phase of the research, it is likely that a very unusual set of features will prove to be the most useful one in identifying languages. It will be of interest to examine the relationships that may exist between those features which are determined to be the greatest contributors to the classification process and basic linguistic units. [Work supported by RADC.]