Jan Nouza

Making Czech Historical Radio Archive Accessible and Searchable for Wide Public

In this paper we describe a complex software platform that is being developed for the automatic transcription and indexation of the Czech Radio archive of spoken documents. The archive contains more than 100.000 hours of audio recordings covering almost ninety years of public broadcasting in the Czech Republic and former Czechoslovakia. The platform is based on modern speech processing technology and includes modules for speech, speaker and language recognition, and tools for multimodal information retrieval. The aim of the project supported by the Czech Ministry of Culture is to make the archive accessible and searchable both for researchers as well as for wide public. After the first project’s year, the key modules have been already implemented and tested on a 27.400-hour subset of the archive. A web-based full-text search engine allows for the demonstration of the project’s current state.

Browsing, indexing and automatic transcription of lectures for distance learning

This paper presents a complex system developed to improve the quality of distance learning by allowing people to browse the content of various (academic) lectures. The system consists of several main modules. The first automatic speech recognition (ASR) module is designed to cope with inflective Czech language and provides time-aligned transcriptions of input audio-visual recordings of lectures. These transcriptions are generated off-line in two recognition passes using speaker adaptation methods and language models mixed from various text sources including transcriptions of broadcast programs, spontaneous telephone talks, web discussions, thesis, etc. Lecture recordings and their transcriptions are then indexed and stored in the database. The next module, client-server web lecture browser, allows to browse or play the indexed content and search in it.

A voice-operated multi-domain telephone information system

This contribution describes the first Czech voice-operated information system running on the public telephone network. The system named InfoCity is a multi-domain inquiry service. It offers a wide range of information useful for citizens of a middle-size city, such as programs of cinemas and theatres, overview of local sport events, timetables of city and regional transport, opening times of public or private institutions. Since March 1999 the InfoCity serves inhabitants and visitors of Liberec, a city with 100000 population. During the first 9 months of its operation, more than 2900 inquiry calls were automatically processed. Although the system uses a fixed dialogue scenario with discrete-utterance input (with 220-item vocabulary), long-term statistics show that a typical call (with 5 to 9 dialogue turns) does not take more than 2 minutes. Under field trial conditions the average recognition score is about 87%.

A city information system operating over the telephone

We present a voice controlled inquiry system designed for operation on the public telephone network. Its goal is to provide residents as well as visitors of a city practical information of various types, such as programs of cinemas and theatres, overview of local sport events, timetables of city and regional transport and opening times of public or private institutions. The InfoCity system, as it is named, operates on a system-driven dialogue platform with an isolated-word style input (using a vocabulary of some 220 words and short phrases) and with a TTS generated output. The InfoCity framework has been adapted for serving the city of Liberec. It has been the first automatic information system of this type developed and tested in Czechia.

Using TTS for fast prototyping of cross-lingual ASR applications

In the paper we propose a method that simplifies initial stages in the development of speech recognition applications that are to be ported to other languages. The method is based on cross-lingual adaptation of the acoustic model. In the search for optimal mapping between the target and original phonetic inventories we utilize data generated in the target language by a high-quality TTS system. The data is analyzed by an ASR module that serves as a partly restricted phoneme recognizer. We demonstrate the method on Czech-to-Polish adaptation of two prototype systems, one aimed at handicapped persons and another prepared for fluent dictation with large vocabulary.

Discrete and fluent voice dictation in czech language

This paper describes two prototypes of voice dictation systems developed for Czech language. The first one has been designed for discrete dictation with the lexicon that includes up to 1 million most frequent Czech words and word-forms. The other is capable of processing fluent speech and it can work with a 100,000-word lexicon in real time on recent high-end PCs. The former has been successfully tested by handicapped persons who cannot enter and edit texts by standard input devices (keyboard and mouse).