The application of XML in speech (I.)

Xml

As XML is widely used in industry, different industries in different fields derive some XML subsets that are applicable in some specific fields. One of the voicexml that will be introduced in this series of articles is the main one. It is very similar to the status of WML. WML is a special language used in wireless internet, and VoiceXML is a special language used in voice browsing. Wireless Internet (WAP) and voice browsing are the two hotspots of network development now, and XML has found its use in them.

The whole article will compare and introduce some important links in the application of XML in speech. In addition to VoiceXML, there are other aspects of XML that are used in speech technology. In this section is mainly the introduction of background knowledge.

Voice Browser, a hotbed of voicexml birth

The phone appears to facilitate communication. The computer appears to be calculated. After entering the post-PC era, the two words "computer" and "network" are getting closer. Nowadays, many people who buy computers are able to reach their goal to surf the Internet. The network has become a shortcut for information exchange and sharing, and people have been "shot down". So that the Internet is also on the agenda. The simpler it is, the more it is favored by the widest range of ordinary consumers. The more widely favored by the most common consumers, the more highly praised by the industry.

There is no doubt that there are more telephones than computers. Walk around the streets of a city and you'll easily find a pay phone. But I still can't imagine the "public computer Pavilion" that is ubiquitous on the street, and the five-dime three-minute internet access. Not only in the city, the phone is not a rare thing, but to buy a computer, many people still have to bite the teeth. Moreover, the advantage of telephones over computers is growing. Personal mobile phone development is in the unprecedented peak period, I believe that the future of single machine is not a myth.

In addition, the telephone is far more affinity than the computer. I believe that many people have a certain tool mentality in the operation of the computer. Not afraid of the original interface has been changed can not be restored, is afraid of not knowing where to modify the desired settings. This is already a friendlier window interface, not to mention the full screen black Unix and so on. And the phone is not the same, only a very limited number of keys, encounter trouble put the microphone a hang again to pick up is. And the key thing is that the handset says "words".

So we say that access to the Internet through the telephone is very promising. Plus now there is a trend to the various forms of access to the interface, such as voice, image and touch, together on the phone to become a veritable PDA (personal digital assistant), we have to be here once again to the development of the phone boast about the future. People will be able to connect to the network at any time and at any place via a "phone" that is readily available, in a straightforward way.

Voice browsing is a new way of browsing the network

Owning a telephone switchboard is not a novelty for a small, scaled company. Telephone users can contact different employees through a number of keystrokes. There is also a popular telephone switchboard function is to use or through the digital keypad free layered directory structure to obtain information. And the voice browsing we're going to be talking about here is similar to the form here. Users mainly send their own information by voice. We can also extend the service to the Web site that provides content services. The contact network used is not necessarily a dedicated telephone network, it can also be the Internet.

What is a voice browser

The speech browser is the input channel with the voice input, the other form is the auxiliary input channel, the translation explains the execution speech markup language, and can produce the speech output the equipment.

The explanation for the above taste, such as crap, originates from the consortium. Like to use a word to describe a person, it is quite a lot of trouble to outline some vague generality. Walking upright with legs and mammals with upper limbs. What the Voice browser wants to achieve is through the most affinity of communication, the information from the network with the most abundant resources through the voice to the user.

With a discerning eye, the design completes a voice browser and a technical barrier to speech recognition, speech comprehension and speech synthesis. When a user wants to make a command out of voice, the voice browser needs to know what the user is saying? It is simpler to command speech recognition. The system requires you to say "mosquito" or "fly". The others are considered illegal input. Complexity involves understanding a piece of speech. For example, when the user says, "What day is today?" "or" Today is the first day of one weeks? "When you know it's the same thing. In this way, users are much more convenient. And this requires speech understanding technology to deal with. Speech synthesis is also essential. When you want to send the browsing information to the user from time to time through the voice. This can be a pre-recorded voice of a natural person, or a TTS speech synthesis system through text-to-speech. It can be very clear that the above to use the technology is required a lot of technology accumulation to achieve, is not a day of work. For the vast majority of design to achieve the development of voice browser will be very natural to adopt some in the voice of the company provides a lot of technical product support. The main focus is on integration.

The main features of the voice browser are very distinct. In many ordinary occasions, through an approximate nature of the dialogue to publish the command, through a different than the HTML two-dimensional browsing mode, the time to collect information linearly. It's just very tempting. In many cases, eyes are not allowed to be supervised at all, as when driving a car. The user is controlled only by a non-visual way.

Web browsers provide a way to jump between different foliage, and voice browsers can jump from one conversation to another. Web browsers provide a way to add and send forms, and a voice browser page can set a user's voice for a single purposeful input. In fact, there is a lot of similarity between voice browsers and HTML foliage browsers. The GRE analogy is the way to say: Voice browser: voicexml::html browser: HTML.