Programme for SCALLA 2001

Wednesday 21st November

9:30 Introduction and Welcome Chair: S.P.Mudur

9:20 - S.P. Mudur - Welcome.

9:30 - Pat Hall – Overview of the conference.

10:10 Localisation needs

Why is localisation important? What computer support is needed for dominant languages? How could computer support save endangered languages?

10:10 - Nick Ostler – Local languages.

11:00 Tea and coffee

11:30 Localisation needs (continued) Chair: Ramakrishnan

11:30 - V Vinay - The Simputer.

12:15 - Vijay Paratap Singh Aditiya– the Honey Bee networks.

13:00 Lunch

14:00 Localisation practices Chair: B.B. Chaudhuri

How is software currently localised, and what are the limitations of current approaches? Skills, tools, and technologies required. The economics of localisation. Localisation in Europe and India. How can we move beyond current practice? Can developments in software technology over the past few decades help us?

14:00 - Reinhard Schaler - Localisation in Europe

14:30 - S.P.Mudur - Localisation in India

15;00 - Pat Hall – Software components and APIs.

15:30 Tea and coffee

16:00 Writing systems, input and output Chair: Pat Hall

How are writing systems represented in the computer? How can writing be input to the computer and output from the computer? Are current methods adequate, are deeper linguistic issues revealed?

16:00 - R.K. Joshi – Writing Systems in India

16:30 - B.B. Chaudhuri - Scripts and OCR for South Asian languages.

17:00 Culture and other factors Chair: Nick Ostler

General Discussion.

We know that some cultural aspects like calendar systems, colour, naming conventions, must be handled. But how about more subtle issues like the way languages are described, the relationship of power between the computer and the human user? Indeed, is the very idea of computers and software and content alien to South Asian cultures?

18:00 End of proceedings for day

Thursday 22nd November

9:30 Language Models Chair: B.N. Patnaik

What descriptions of language are possible, and appropriate? Are there significant differences between languages across South Asia, and between South Asia and Europe? How do you set about describing a language, when it is written and only has a few hundred speakers?

9:30 - Boyd Michailovsky - Creating and exploiting a computerized speech corpus for lexicography andlinguistic research.

10:15 - Harold Somers - Developing linguistic resources from corpus material.

11:00 Teach Coffee

11:30 Translation and Language generation

Localising software and content can be viewed as a translation task, taking text in one language (the various messages in the software, the text of the content) and producing equivalent text in the target language. However, it can also be viewed as communication from a machine which holds an abstract representation of the knowledge that a person needs to that person, able to output that knowledge through the generation of appropriate natural language sentences.

11:30 - Donia Scott - Multilingual Natural Language Generation.

13:00 Lunch

14:00 Lexicography

The present scenario of dictionaries in countries like India or Bangladesh, in comparison to the status of dictionaries in Europe or America. The use of dictionaries. How language corpora can bring a total change in the outlook of dictionary development and use. How should we develop more advanced and authentic dictionaries (both monolingual and bilingual)?

14:00 - Rajeev Sangal and Dipti Sharma - the free electronic English-Hindi dictionary

14:45 - Durgesh Rao – translating from South Asia languages into English.

15:30 Tea and Coffee

16:00 Lexicography (continued)

16:00 - Udaya Narayana Singh and B. Mallikarjun – making a traditional dictionary into an electronic lexicon.

16:45 - Niladri Sekhar Das - The contribution of language corpora on the development of dictionaries.

18:00 End of proceedings for the day.

19:00 Conference Dinner

Friday 23rd November

9:30 Speech and literacy

With literacy in South Asia at just under 50%, there is a great need for computers to work in speech. At its simplest this can be achieved for fixed messages and content by pre-recording these, but we need more than this. We need to be able to produce speech in many languages and dialects. And we need to enable non-literate people to create new content for access by other persons.

9:30 - Asoke Kumar Dutta – Disbursing spoken language technology in regional dialects.

10:00 - Gautum Sengupta - Voice-enabled Machine Readable Dictionaries for Indian Languages

10:30 - Roger Tucker – a pure speech personal digital assistant.

11:00 – Tea and coffee

11:30 – General Discussion, summing up, the future

What have we learnt during the past three days? What would be appropriate for the next two SCALLA conferences?

13:00 Lunch and End of conference.