Innovation: Poland's technologists talk their way to the top


Poland's advanced technology market is picking up pace. Polish programmers, network engineers, scientists and technology experts are in demand and as Joanna Najfeld of Polish Radio reports - in one case they're talking their way to the top.

One of Poland's recent successes in the field of advanced technologies took place at Blizzard Challenge 2007- an international speech processing competition hosted by Edinburgh University. IVONA, a Polish speech synthesizer, got the best score of all sixteen research teams from the world's top university centers and businesses.

We asked IVO Software, the company which developed IVONA, to tell us about their innovative product. They decided IVONA would best speak for itself...

[synthesized female voice speaking] 'The IVONA speech synthesizer is an innovative software that changes text into speech. IVONA can read any kind of text just as well as a professional human lector. The quality and natural-sounding voice of IVONA received one of the highest scores for voice quality at the prestigious scientific events Blizzard Challenge 2007 and 2006. It was discovered by two scientists from IVO Software company, which is now the fastest developing Polish provider of speech technology.'

Speech synthesizers are more widespread in your everyday life than you would think. Whenever you send a text message from your mobile to a landline phone, it is read out to the recipient by a speech synthesizer. This kind of software is also used on websites, in education, enterprise and entertainment, whenever a written message needs to be converted into speech signal. With a speech synthesizer, you can now create an audio file of any written document - your favourite book or the morning paper - to listen to it while jogging or driving to work. Needless to say, speech processing technology is absolutely irreplaceable for blind and visually impaired people, such as Zbigniew Drzazga:

'The technology allows blind people to be more independent than ever before, more active in social life, gives them better access to the job market, more equal chance to education. I can benefit from online banking, online shopping, internet sites to read news, weather forecast and other things. Blind people can communicate via e-mail, or chat. This is possible because computer technology is developing so quickly.'

Improving the quality of the speech produced is not just art for art's sake. Zbigniew Drzazga again:

'It is not important if you use the Iinternet to do online banking or to read a weather forecast, but if you read a book, it is very important to have intonation, because if the quality of speech is bad, you are not able to listen to the text for a long time, because it is so boring to hear a voice of a bad quality.'

From the scientific point of view, speech synthesis is part of a larger domain of speech processing, explains Artur Janicki of the Warsaw Technical University. Just as you can make the computer read a text, you can also make it take dictation, that is write down what is being said. The newest technology also allows you to analyze the speech for additional information...

'We can try to estimate what the emotional state of a speaker is when uttering a sentence. For example when a company wants to analyze what the mood of the client is, we can try to figure out whether the speaker was calm, was satisfied after talking or got nervous and was not satisfied with the service.'

Poland has a history of achievements in the area of speech synthesis, continues speech processing expert Artur Janicki:

'I can recall speech synthesizer from the 1980s, which was called Kubuz, especially useful for visually impaired people. This technique has evolved. Earlier, we were using simple techniques, whose aim was only to produce speech which is intelligible, so that we can understand, what is said. But now the focus is to make speech sound natural and this is the biggest challenge now.'

We wish speech synthesis technology to continue to develop and flourish, but still, we do hope, it will never get to replace human radio announcers.