Towards Improved Assessment of Phonotactic Information for Automatic Language Identification

Language identification technology is widely used in the domains of machine learning and text mining. Many researchers have achieved excellent results on a few selected European languages. However, the majority of African and Asian languages remain untested. The primary objective of this research is to evaluate the performance of our new n‑gram based language identification algorithm on 68 written languages used in the European, African and Asian regions. The secondary objective is to evaluate how n‑gram orders and a mix n‑gram model affect the relative performance and accuracy of language identification. The n-gram based algorithm used in this paper does not depend on the n‑gram frequency. Instead, the algorithm is based on a Boolean method to determine the output of matching target n‑grams to training n‑grams. The algorithm is designed to automatically detect the language, script and character encoding scheme of a written text. It is important to identify these three properties due to the reason that a language can be written in different types of scripts and encoded with different types of character encoding schemes. The experimental results show that in one test the algorithm achieved up to 99.59% correct identification rate on selected languages. The results also show that the performance of language identification can be improved by using a mix n‑gram model of bigram and trigram. The mix n-gram model consumed less disk space and computing time, compared to a trigram model. DOI: 10.4038/icter.v2i2.1385 The International Journal on Advances in ICT for Emerging Regions 2009 02 (02): 21-28

This study proposes a novel deep neural network (DNN) based approach to language identification (LID) for the NIST 2015 Language Recognition (LRE) i-Vector Machine Learning Challenge. State-of-the-art DNN based LID systems utilize large amounts of labeled training data. The 2015 LRE i-Vector Machine Learning Challenge limits the access to only ready-to-use i-Vectors for LID system training and testing. This poses unique challenges in designing DNN based LID systems, since optimized front-ends and network architectures can no longer be used. We propose to use the training i-Vectors to train an initial DNN for LID. Next, we present a novel strategy to use this initial DNN to estimate out-of-set language labels from the development data. The final DNN for LID is trained using the original training data, and the estimated out-of-set language data. We show that augmenting the training set with out-of-set labels leads to significant improvement in the LID performance. Our approach obtains very competitive costs (defined by NIST) of 26.56, and 25.98 respectively, on the progress and evaluation subsets of the challenge. Since the amount of training data is very limited (300 i-Vectors per language), this study outlines a successful recipe for DNN based LID using very limited resources.

The classical language identification architecture would require a collection of languages independent text and speech information for training by the system before it can identify the languages correctly. This paper also address language identification framework however with data has been downsized considerably from general language identification architecture. The system goal is to identify the type of language being spoken to the system based on a series of trained speech with sound file features and without any language text data or lexical knowledge of the spoken language. The system is also expected to be able to be deployed in mobile platform in future. This paper is specifically about measuring the performance optimisation of audio filters on a CNN model integration for the language identification system. There are several metric to gauge the performance identification system for a classification problem. Precision, recall and F1 Scores is presented for the performance evaluation with different combination of filters together with CNN model as the framework of the language identification system. The goal is not to get the best filter for noise, instead to identify the filter that is a good fit to develop language model with environmental noise for a robust language identification system. Our experiments manage to identify the best combination of filters to increase the accuracy of language identification using short speech. This resulting us to modify our pre-processing phase in the overall language identification system.

Human and computer recognition of regional accents and ethnic groups from British English speech

This paper describes the development of a Language Identification (LID) system oriented to robust Multilingual Speech Recognition in the Basque context where coexist three languages: Basque, Spanish and French. The LID system is integrated in GorUP, a Semantic Speech Recognition system for industrial complex environments described in Part I. The work presents hybrid strategies for LID, based on the selection of system elements by several classifiers (Support Vector Machines and Multilayer Perceptron) and Discriminant Analysis improved with robust regularized covariance matrix estimation methods oriented to under-resourced languages and stochastic methods for speech recognition tasks (Hidden Markov Models and n-grams). The LID tool manages the main elements of the Automatic Speech Recognition system (Acoustic Phonetic Decoder, Language Model and Lexicons).

This article addresses language identification at the word level in Indian social media corpora taken from Facebook, Twitter and WhatsApp posts that exhibit code-mixing between English-Hindi, English-Bengali, as well as a blend of both language pairs. Code-mixing is a fusion of multiple languages previously mainly associated with spoken language, but which social media users also deploy when communicating in ways that tend to be rather casual. The coarse nature of code-mixed social media text makes language identification challenging. Here, the performance of deep learning on this task is compared to feature-based learning, with two Recursive Neural Network techniques, Long Short Term Memory (LSTM) and bidirectional LSTM, being contrasted to a Conditional Random Fields (CRF) classifier. The results show the deep learners outscoring the CRF, with the bidirectional LSTM demonstrating the best language identification performance.

It is significant to highlight that the demand of multilingual systems in human-computer interfaces is growing and automatic language identification (LID) is becoming increasingly important for the speech community as a fundamental tool to enrich theses systems. Research in this field has been active for several years. LID becomes also a useful resource in Multilingual Security System (MSS) providing methodologies to improve the robustness and the effectiveness of these systems. The main goal of our project is the development of Multilingual LVCSR systems in the Basque context. Nowadays, our work is oriented to the use of Basque in internet. The present work describes the development of a Language Identification (LID) system oriented to robust Multilingual Speech Recognition for under-resourced languages in the Basque context.

Language identification is the task of automatically determining the identity of a language conveyed by a spoken segment. It has a profound impact on the multilingual interoperability of an intelligent speech system. Despite language identification attaining high accuracy on medium or long utterances (>3 s), the performance on short utterances (<=1s) is still far from satisfactory. We propose a BERT-based language identification system (BERT-LID) to improve language identification performance, especially on short-duration speech segments. We extend the original BERT model by taking the phonetic posteriorgrams (PPG) derived from the front-end phone recognizer as input. Then we deployed the optimal deep classifier followed by it for language identification. Our BERT-LID model can improve the baseline accuracy by about 6.5% on long-segment identification and 19.9% on short-segment identification, demonstrating our BERT-LID’s effectiveness to language identification.

This paper proposes a new phone lattice based method for automatic language recognition from speech data. By using phone lattices some approximations usually made by language identification (LID) systems relying on phonotactic constraints to simplify the training and decoding processes can be avoided. We demonstrate the use of phone lattices both in training and testing significantly improves the accuracy of a phonotactically based LID system. Performance is further enhanced by using a neural network to combine the results of multiple phone recognizers. Using three phone recognizers with context independent phone models, the system achieves an equal error rate of 2.7% on the Eval03 NIST detection test (30s segment, primary condition) with an overall decoding process that runs faster than real-time (0.5xRT).

Language identification for handwritten document images using a shape codebook

Computing sparse representation (SR) over an exemplar dictionary is time consuming and computationally expensive for large dictionary size. This also requires huge memory requirement for saving the dictionary. In order to reduce the latency and to achieve some diversity, ensemble of exemplar dictionary based language identification (LID) system is explored. The full diversity can be obtained if each of the exemplar dictionary contains only one feature vector from each of the language class. To achieve full diversity, a large number of multiple dictionaries are required; thus needs to compute SR for a particular test utterance as many times. The other solution to reduce the latency is to use a learned dictionary. The dictionary may contain unequal number of dictionary atoms and it is not guaranteed that each language class information is present. It totally depends upon the number of data and its variations. Motivated by this, language specific dictionary is learned, and then concatenated to form a single learned dictionary. Furthermore, to overcome the problem of ensemble exemplar dictionary based LID system, we investigated the ensemble of learned-exemplar dictionary based LID system. The proposed approach achieves the same diversity and latency as that of ensemble exemplar dictionary with reduced number of learned dictionaries. The proposed techniques are applied on two spoken utterance representations: the i-vector and the JFA latent vector. The experiments are performed on 2007 NIST LRE, 2009 NIST LRE and AP17-OLR datasets in closed set condition.

International audience

Utilising the known language of origin of a name can be useful when predicting the pronunciation of the name. When this language is not known, automatic language identification (LID) can be used to influence which language-specific grapheme-to-phoneme (G2P) predictor is triggered to produce a pronunciation for the name. We investigate the implications when both the LID system and the G2P system generate errors: what influence does this have on a resulting speech recognition system? We experiment with different approaches to LID-based dictionary creation and report on results in four South African languages: Afrikaans, English, Sesotho and isiZulu.

This chapter describes the types of information that can be used to characterize spoken languages. Automatic spoken language identification (LID) systems, which are tasked with determining the identity of the language of speech samples, can utilize a variety of information sources in order to distinguish among languages. In this chapter, we first define what we mean by a language (as opposed to a dialect). We then describe some of the language collections that have been used to investigate spoken language identification, followed by discussion of the types of features that have been or could be utilized by automatic systems and people. In general, approaches used by people and machines differ, perhaps sufficiently to suggest building a partnership between human and machine. We finish with a discussion of the conditions under which textual materials could be used to augment our ability to characterize a spoken language.

An audio utterance can be identified as being spoken in a particular language by using automatic language identification (LID). Each language has its own phoneme set. Hence combination of these phonemes governed by phonotactics will help in distinguishing languages. In this work, we propose an automatic language identification system utilizing features derived from non-uniform speech regions to represent phonotac-tic differences among 4 Indian languages, namely Malayalam, Marathi, Assamese, and Kannada. For this, broad phoneme labels, namely approximant (A), closure (C), fricatives (F), nasals (N), plosive/stop (P), voiced stop (B), vowels (V), and silence (S) are obtained automatically by a broad phoneme classifier (BPC). It is a DNN-based classifier which uses hand-crafted features and Mel-frequency cepstral coefficients (MFCC). In order to automatically segment speech to smaller regions, first it is chopped at every silence regions using the labels obtained from BPC. Later, it is split again at the end of each vowel. Hence, small non-uniform regions are obtained which contain phoneme combinations that may be specific to the language of the utterance. From each region, only a fixed number of frames containing certain combination of phonemes are selected. A DNN classifier is trained using 13-dimensional MFCC features of 12 fixed frames of non-uniform regions for performing LID. An average accuracy of 97.03% is obtained for test utterances of 10 sec duration belonging to 4 languages.