A variety of approaches to language identification, based on (a) acoustic features, (b) broad-category segmentation, and (c) fine phonetic classification, are introduced. These approaches are evaluated in terms of their ability to distinguish between English and Japanese utterances spoken over a telephone channel. It is found that the best performance (86.3 % accurate classification of utterances with a mean length of 13.4 sec) is obtained when fine phonetic features are employed. In addition, the results show the importance of discriminatory training rather than likelihood estimation. 1. INTRODUCTION As developments in telecommunications and long-distance travel cause national borders to become increasingly transparent, the ability to identify which language is being spoken is growing in importance. The utility of tasks such as directory assistance or automatic translation is, for instance, improved substantially by the availability of a means of identifying which language is being s...

Automatic Language Identification (ALI) is the problem of automatically identifying the language of an utterance through the use of a computer. In 1977, House and Neuburg proposed an approach to ALI which focused on the phonotactic constraints of different languages. Their work suggested that simple language models could be used effectively for language identification if an accurate phonetic representation of an utterance could be obtained from the acoustic signal. Our research utilizes House and Neuburg's ideas as the starting point for a new segment-based approach to ALI. To develop a solid theoretical basis for the design of an ALI system, a formal probabilistic framework has been developed. This framework uses House and Neuburg's ideas as its foundation but also utilizes additional information that may be useful for ALI. Specifically, phonotactic, acoustic and prosodic information are all incorporated into the framework which provides the structure for the segment-based system. To ...

This paper presents an analysis of the phonemic language identification system introduced in [5], now extended to recognize German in addition to English and Japanese. In this system language identification is based on features derived from a superset of phonemes of all three languages. As we increase the number of languages, the need to reduce the feature space becomes apparent. Practical analysis of single-feature statistics in conjunction with linguistic knowledge leads to 90% reduction of the feature space with only a 5% loss in performance. Thus, the system discriminates between Japanese and English with 84.1% accuracy based on only 15 features compared to 84.6% based on the complete set of 318 phonemic features (or 83.6% using 333 broad-category features [4]). Results indicate that a language identification system may be designed based on linguistic knowledge and then implemented with a neural network of appropriate complexity. 1. INTRODUCTION In [5] we introduced a language-ide...

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 OGI Multi-Language Telephone Speech Corpus and the results are competetive with other current LID systems. The results have also indicated that, while the phontotactic information of a spoken utterace 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.

There has been renewed interest in the field of automatic language identification over the past two years. The advent of a public-domain ten-language corpus of telephone speech has made the evaluation of different approaches to automatic language identification feasible. In an effort to provide benchmarks for evaluating machine performance, we conducted perceptual experiments on 1-, 2-, 4- and 6-second excerpts of telephone speech excised from spontaneous speech utterances in this corpus. The subject population consisted of 10 native speakers of English and 2 speakers from each of the remaining 9 languages. Statistical analyses of our results indicate that duration of the excerpt, familiarity with the language, and number of languages known are important factors affecting a subject's performance on the identification task. 1. INTRODUCTION Automatic language identification, the problem of recognizing what language is being spoken, is a challenging research problem with important real-w...

Introduction 1.1 The Problem Automatic language identification (language ID for short) is the problem of identifying the language being spoken from a sample of speech by an unknown speaker. As with speech recognition, humans are the most accurate language identification systems in the world today. Within seconds of hearing speech, people are able to determine whether it is a language they know. If it is a language with which they are not familiar, they often can make subjective judgments as to its similarity to a language they know, e.g., "sounds like German". Languages have characteristic sound patterns; they are described subjectively as "singsong", "rhythmic", "guttural", "nasal" etc. Languages differ in the inventory of phonological units (speech sound categories) used to produce words, the frequency of occurrence of these units, and the order in which they occur in words. The presence of individual sounds, such as the "clicks" found in some sub-Saharan African la

xii 1 Introduction 1 1.1 Background : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 2 1.1.1 Nature of the Problem : : : : : : : : : : : : : : : : : : : : : : : : : : 2 1.1.2 The Difficulties: Challenges to LID : : : : : : : : : : : : : : : : : : : 5 1.2 Related Work : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 7 1.2.1 Early Work: 1973--1992 : : : : : : : : : : : : : : : : : : : : : : : : : 7 1.2.2 Current Activities: 1992--present : : : : : : : : : : : : : : : : : : : : 9 1.2.3 The Problems : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 11 1.3 An Approach to Language Identification based on language-dependent phone recognition. : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 12 1.3.1 Finding a Good Modeling Unit : : : : : : : : : : : : : : : : : : : : : 12 1.3.2 The Baseline System : : : : : : : : : : : : : : : : : : : : : : : : : : : 13 1.3.3 Contributions: Methods Proposed to Improve the Baseline...

he multilingual problems just 282 Chapter 8: Multilinguality identified, the only one that might possibly be treated with a character-oriented model is that of language identification. The remainder trade in an essential way on equivalences, or near equivalences, among words, sentences, and texts mediated through their meaning. Language processing of this kind is notoriously difficult and it behooves us to start by considering, however cursorily, why this is. We will do this in the context of translation, though what we say is true for the most part of the other tasks mentioned. The question of why translation should have been so successful in resisting the most determined efforts to automate it for close to forty years is complex and sometimes quite technical. But it is not a mystery. The basic problems have long been known and, the most important thing that has been learnt about them recently is that they are more severe and more widespread than was first thoug

Automatic language identification (LID) continues to play an integral part in many multilingual speech applications. The most widespread approach to LID is the phonotactic approach, which performs language classification based on the probabilities of phone sequences extracted from the test signal. These probabilities are typically computed using statistical phone n-gram models. In this paper we investigate the approximation of these standard n-gram models by mixed-memory Markov models with application to both a phone-based and an articulatory feature-based LID system. We demonstrate significant improvements in accuracy with a substantially reduced set of parameters on a 10-way language identification task.

A neural network based language identification system is described, which uses language independent phoneme clusters as speech units to recognize the language spoken by native speakers over the telephone. We extend our previous work comparing phoneme-cluster and phoneme based approaches to language identification [1]. By creating a new speech unit valid across all languages in a theoretically motivated manner, we circumvent problems that are associated with fine phonemic modelling such as high complexity [4], extensive training requirements [2], and the linguistically arbitrary reduction to subsets of phonemes [4]. A common set of speech units across languages allows us to automatically derive discriminating sequences of any length and theoretically estimate the language identification error. We demonstrate our implemented system for German vs. English on the OGI-TS database. 1. Introduction We view automatic language identification as the tagging of an incoming utterance with the co...