Automated Lip Reading (ALR) is a software technology developed by speech recogni tion expert Frank Hubner.

A video image of a person talking can be analysed by t he software. The shapes made by the lips can be examined and then turned into so unds. The sounds are compared to a dictionary to create matches to the words bei ng spoken Imagine moving your lips silently to a computer and watching the unspoken uttera nces roll out on the screen - the ability to control computers without making a sound or touch-typing. This illustrates one of the main benefits of computer-based lip-re ading. This technology invites the possibility of recognizing speech based on the move ment the human lips. A general overview of such computer-based lip reading approaches be presented in this talk followed by a detailed description of the motion technique to recognize utterance without evaluating the sound signals. Potential applications for lip-reading technologies include human computer inter face (HCI) for mobility-impaired users, defense applications that require voice-less commu nication in-vehicle systems, and improvement of speech-based computer control in noise-fi lled environments. Karlsruhe Institute of Technology (KIT) has come up with a way that makes room f or soundless calling, where it is on exhibition at CeBIT 2010. To sum it up in t heir own words, here it is below. The Institute for Anthropomatics has developed a technology that enables people to speak soundlessly and still to be understood by the conversation partner. Thi s technology is based on the principle of electromyography, that is the acquisit ion and recording of electrical potentials generated by muscle activity. This mu scle activity is measured in the face. Four applications of this technology will be exhibited. An example is soundless calling. The user can speak into the phon e soundlessly, but is still understood by the conversation partner on the other end of the line. As a result, it is possible to communicate in silent environmen ts, at the cinema or theater, without disturbing others. Another field of use is the transmission of confidential information. is hardly a new form of communication: it is a primary tool of the deaf communit y, not to mention a useful skill for transmitting messages across a loud party o r over the heads of children. A new form of technology, however, is attempting t o mechanize this age-old method of communication. Automatic lip reading, also kn own as automatic speech reading, is a growing branch of speech recognition techn ology. In theory, by monitoring a speaker's lip movements and other related elem ents, specially designed computer programs can interpret verbal messages even wh en noise interference or other obstacles prohibit a human voice from being prope rly heard. Usually, automatic lip reading requires a video of the speaker. The shapes and r hythms formed by the lips are often the most important factors in interpretation , but a variety of other movements can help to decipher the speaker's message, a s well. Facial expressions and movements of the head can be determining factors. If any useful audio is available, the automatic lip reading process is signific antly facilitated, because context clues from the audible portion can suggest in

audible words or fragments of speech. The development of automatic lip reading technology rests on the ability to iden tify isolated words that use the same lip movements every time they are spoken. Lip Geometry Estimation (LGE) is one of the more advanced computer systems in pl ace that monitors and interprets such patterns. Scientists have developed a nota ble automatic lip reading and speech recognition system for the Dutch language, but in general, the field still has plenty of room for progress. Automatic lip reading has received a recent surge of interest and attention due to its implementation on archived home videos of Adolf Hitler. Filmed by Eva Bra un during the war, the once-silent films came to life when speech recognition te chnology was applied to the footage. Whether or not the interpretations are enti rely accurate, the automatic lip reading program has provided a glimpse into a h istorical figure that will no doubt create continued interest in the field of au tomatic lip reading. Computers that can read lips are already in development but this is the first ti me they have been 'taught' to recognise different languages. The discovery could have practical uses for deaf people, for law enforcement agencies, and in noisy environments. Led by Stephen Cox and Jake Newman of UEA's School of Computing Sciences, the gr oundbreaking research will be presented at a major conference in Taiwan on Wedne sday April 22. The technology was developed by statistical modelling of the lip motions made by a group of 23 bilingual and trilingual speakers. The system was able to identif y which language was spoken by an individual speaker with very high accuracy. Th ese languages included English, French, German, Arabic, Mandarin, Cantonese, Ita lian, Polish and Russian. "This is an exciting advance in automatic lip-reading technology and the first s cientific confirmation of something we already intuitively suspected - that when people speak different languages, they use different mouth shapes in different s equences," said Prof Cox. "For example, we found frequent 'lip-rounding' among French speakers and more pr ominent tongue movements among Arabic speakers."