Tsuyoshi Morimoto

Automatic acquisition of probabilistic dialogue models

In the work described, we automatically deduce dialogue structures from a corpus with probabilistic methods. Each utterance in the corpus is annotated with a speaker label and an utterance type called IFT (Illocutionary Force Type). We use an ergodic HMM (hidden Markov model) and the ALERGIA algorithm, an algorithm for learning probabilistic automata by means of state merging, to model the speaker-IFT sequences. Our experiments successfully extract typical dialogue structures such as turn taking and speech act sequencing.

A spoken language translation system: SL-trans2

1. I n t r o d u c t i o n An automatic telephone interpreting system wil l u n d o u b t e d l y be u s e d to o v e r c o m e c o m m u n i c a t i o n b a r r i e r s b e t w e e n people speaking different languages. Recently, great interest has been growing in this area [Saitoh88], [Waibel-91l, [Kitano-91], [Roe-92]. SLTRANS2 .1~ is an experimental system developed at ATR, which t rans la tes Japanese speech to English speech. It is composed of three basic components: speech recognition, translation and speech synthesis . This paper in t roduces the s y s t e m wi th e m p h a s i s on the t r a n s l a t i o n component. The discourse domain is a dialogue c o n c e r n i n g an i n t e r n a t i o n a l c o n f e r e n c e registrat ion. The d is t inc t ive f ea tu re s of the system are as follows. (1) Japanese continuous speech input can be recognized with h igh accuracy . Moreover , speaker independent recognition using speaker adaptation technique has been developed. (2) Var ious express ions pecul ia r to spoken language can be accepted and translated properly into the target language. In Japanese, the style of spoken sentences is general ly quite different from that of written texts. Spoken utterances are f ragmentary and include the speakers intention directly or indirectly. The system extracts the intent ion and then t ranscr ibes it to a proper expression in the target language. (3) Linguistic knowledge sources necessary for the translation are defined declaratively to the e x t e n t . Such d e f i n i t i o n i m p r o v e s h i g h modularity, readability and easy maintenance of knowledge description. In the next section, the system is overviewed and a brief description of the speech recognition

Schema method: a framework for correcting grammatically ill-formed input

The schema method is a framework for correcting grammatically ill-formed input. In a natural language processing system ill-formed input cannot be overlooked. A computer assisted instruction (CAI) system, in particular, needs to show the users errors. This framework diagnoses ill-formed input, corrects it and explains the error, if an input is ill-formed. The framework recognizes a sentence at two steps: first parses weak grammar, and then strongly filters the parsed sentence. When it is known what sentences are passed by the filter, it can be used even if it is imperfect. As the strong filter, a new method is used: an interpretation schema and an interpretation rule. An interpretation schema collects input information schemata and then an interpretation rule judges whether the collected schemata are correct or incorrect. This approach overcomes the problem of relaxation control, the major drawback of the previous syntactically-oriented methods, and is also more efficient.

Linguistic constraints for continuous speech recognition in goal-directed dialogue

The application of language models to continuous speech recognition, and the kinds of linguistic constraints important for improving speech recognition accuracy are discussed. First, the authors conducted an experiment to see how inter-phrase grammar contributes to the improvement of sentence speech recognition rates. The inter-phrase grammar was developed according to a dialogue database of an international conference secretarial service containing about a half-million words. It is confirmed that syntax is a very effective constraint for continuous speech recognition. Moreover, the predictive two-level LR parsing method is much better than the phrase lattice parsing method. The kinds of constraints that should, or can be, used to suppress each remaining error are discussed by referring to an example.<<ETX>>

Restructuring tagged corpora with morpheme adjustment rules

A part-of-speech tagged corpus is a very important knowledge source for natural language processing researchers. Today, several part-of-speech tagged corpora are readily available for research use. However, because there is wide diversity of morphological information systems (word-segmentation, part-of-speech system, etc.), it is difficult to use tagged corpora with an incompatible morphological information system. This paper proposes a method of converting tagged corpora from one morpheme system to another.

Statistical method of recognizing local cohesion in spoken dialogues

This paper presents a method for automatically recognizing local cohesion between utterances, which is one of the discourse structures in task-oriented spoken dialogues. More specifically we can automatically acquire discourse knowledge from an annotated corpus with local cohesion. In this paper we focus on speech act type-based local cohesion. The presented method consists of two steps 1) identifying the speech act expressions in an utterance and 2) calculating the plausibility of local cohesion between the speech act expressions by using the dialogue corpus annotated with local cohesion. We present two methods of interpolating the plausibility of local cohesion based on surface information on utterances. The presented method has obtained a 93% accuracy for closed data and a 78% accuracy for open data in recognizing a pair of utterances with local cohesion.