This is the extended version of a Comment submitted to Physical Review Letters. I first point out the inappropriateness of publishing a Letter unrelated to physics. Next, I give experimental results showing that the technique used in the Letter is 3 times worse and 17 times slower than a simple baseline. And finally, I review the literature, showing that the ideas of the Letter are not novel. I conclude by suggesting that Physical Review Letters should not publish Letters unrelated to physics.

This paper describes our work in developing multilingual (Swedish and English) speech recognition systems in the ATIS domain. The acoustic component of the multilingual systems is realized through sharing Gaussian codebooks across Swedish and English allophones. The language model (LM) components are constructed by training a statistical bigram model, with a common backoff node, on bilingual texts, and by combining two monolingual LMs into a probabilistic finite state grammar. This system uses a single decoder for Swedish and English sentences, and is capable of recognizing sentences with words from both languages. Preliminary experiments show that sharing acoustic models across the two languages has not resulted in improved performance, while sharing a backoff node at the LM component provides flexibility and ease in recognizing bilingual sentences at the expense of a slight increase in word error rate in some cases. As a by-product, the bilingual decoder also achieves good performance on language identification (LID). 1.

In this paper we explore the use of lexical information for language identification (LID). Our reference LID system uses language-dependent acoustic phone models and phone-based bigram language models. For each language, lexical information is introduced by augmenting the phone vocabulary with the N most frequent words in the training data. Combined phone and word bigram models are used to provide linguistic constraints during acoustic decoding. Experiments were carried out on a 4-language telephone speech corpus. Using lexical information achieves a relative error reduction of about 20% on spontaneous and read speech compared to the reference phone-based system. Identification rates of 92%, 96% and 99% are achieved for spontaneous, read and task-specific speech segments respectively, with prior speech detection. 1. INTRODUCTION Many state-of-the-art language identification (LID) systems exploit phone-based acoustic and (or) phonotactic scores [7]. Training generally consists of desi...