If You Don’t Do It, Somebody Else Will

This year has been one of dramatic strides for Philips, as they became one of the world’s leading companies in the speech recognition industry, particularly in the telephony market. Obviously, the person in charge of all this is generally busy, but we were able to get Paul Celen to sit long enough to discuss the speech industry with us, and Philips’ role in it both now and in the near future. Speech Technology: Is speech recognition going to really make a dramatic visible change in how people communicate with machines? Paul Celen: Yes, I think speech will do that. Speech is used in numerous applications, such as command and control, telephony and high volume electronics. Suppose you could talk to your television, DVD, video recorder, or any other device you can imagine? It will come and it will change the overall results in the way we use machines. ST: What are the benefits of speech enabled applications? PC: That depends on what people expect out of speech recognition. For instance, in the case of directory assistance, providers see an immediate cost reduction, or increase in the value of their services by making it more available to people in a cost-effective way. They also relieve the stress in their call centers by automating a number of applications that are easy to work with. This frees up time to allow employees to work on more complex matters. So there is an immediate value. ST: How is this going to improve communications? PC: As an example, lets discuss what we will use the phone for. Some people think it is just for phoning. But some companies like Nokia and Ericson think that the mobile phone will be an information device that will need a bigger display, or a little screen and faster technology. With the new technology and the higher speed of the technology, I think we will be able to send information through our mobile phones. Speech will help here. The watt systems are there, so that you can talk to the Internet and receive information over your mobile phone. ST: What are the biggest obstacles for the acceptance of speech? PC: You have to convince people that there is still some risk in implementing it. Now that goes together with expectation. I repeatedly say that people sometimes believe that Star Trek is here. And it isn't. Let's take the example of a speech portal and navigation system combined with a service offer. I ask a lot of people, 'what is the economic value of it? Is it worth doing it?' If you go to a call center, everybody knows that maybe the call completion rate is between 93% and 97%. Is that acceptable compared to a call completion rate over voice? Should we improve that or not improve that? Now then you can ask yourself 'if it would be 100%, then everybody would buy it, wouldn't they,' and that's true, of course. You can train a system. For instance, if you have a stand-alone application, like a call center application, you can train a system in such a manner that it can achieve very high figures, and then I think the economical value is proven. ST: Are we at a point now where the technical issues and challenges are met? PC: I think it's more complex than that. I think we believe that the technology is there. When you talk to technical people within the different companies, they are convinced that the technology is valid -- it works, the accuracy is there, as well as the articulation. There are enough tools that maybe we can improve everything, but integration is easier. Everything is there. ST: What are the marketing challenges now? PC: I see still a lot of hesitation. People that want to wait until they see that it is working, but at least we have a few people who take the initiative. The interesting thing is, if you look at trying to sell services to a large amount of the public, lets say 250 million Americans, if you look at the penetration rates of the PC or Internet, you are at roughly, depending on the measurements, 15 - 17% maximum. If you look at the penetration of mobile telephony, you are at 70%. So what I'm trying to tell people is 'you better sell your services by mobile telephony, because you can reach a big number of people.' Now, the issue is whether or not the telephony companies themselves have to do that, or whether they should be helped by marketing, or by content providers. ST: Where is the growth in the speech industry going to come from? PC: If you look at the consumer dictation market, or the desktop, you have quite a volume already established. There is a new report from Frost & Sullivan, I don't have the figures in my head, but I think it's still growing by the rate of about 30%. I think that is because the desktop market in the United States has already been established for 2 or 3 years. At the moment, with the exception of a few call center installations, the network isn't existing in the United States for the telephony market. So there the growth rate will be much higher. ST: Philips is a leader in the European market, but there are some companies in North America that seem to be a little ahead of Philips in the speech industry. What is Philips' strategy to close this gap in the North American market? PC:With the VCS customers included, we have become a market leader with the number of licenses sold, even numbers of companies installed. We don't make a lot of noise about it, but maybe we should do that. We came into the market in September 98, which was reasonably late. Being on the market for just a year now, people are starting to know us. We are in the upper numbers on the name recognition order, the brand name. Maybe we're not number one yet, but we will come close. I think on the technological side we have a clear advantage. There is nobody else that has the dialog modules that we have. We have modules that are so good that normal conversations can be held. That is a clear advantage. It is certainly present in Europe and we will also bring it to the United States. Philips in 1999
In May in the United States, Philips acquired Dallas-based Voice Control Systems, a leading speech software platform provider. In Italy, Philips partnered with Omnitel to produce Omnitel 2000, the world’s largest net portal and the first that allows access to a large number of services via the telephone, as well as the Internet, using natural language technology. In September in Portugal, Philips and Portugal Telecom announced they would equip more than 1,000 ports in the Telecom’s call center operations, speech enabling services from directory assistant to calling card validation and call completion. On the dictation side, the company recently introduced the Philips SpeechPro, designed for several specialties in the legal and medical markets and unveiled a line of software and hardware products called Philips FreeSpeech2000. About Paul Celen
Paul P. Celen, 52, joined Philips Speech Processing in February 1998 and currently serves as the company’s Chief Operating Officer. In his role as COO, Celen is responsible for the operating activities of Philips Speech Processing and its involvement in domestic and international partnerships, alliances and joint venture agreements. In addition, Celen is responsible for worldwide sales, marketing, engineering and product development. Celen, a native of Belgium, has more that 20 years of experience in general management, marketing, sales and business development within the IT, medical, chemical and industrial markets. Celen has extensive expertise in directing and streamlining major corporate development programs and has successfully implemented strategic activities at ICL plc (a unit of Fujitsu) Ets. Van Der Heyden and Mines, Minerals et Metaux. Celen has a Ph.D. in Science from the State University of Ghent, an MBA from the University of Antwerp, associated with Boston University, and holds several prestigious degrees from global institutions.
SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues
Companies and Suppliers Mentioned