Navigation Links
Doing what the brain does -- how computers learn to listen
Date:8/14/2009

Researchers at the Leipzig Max Planck Institute for Human Cognitive and Brain Sciences and the Wellcome Trust Centre for Neuroimaging in London have now developed a mathematical model which could significantly improve the automatic recognition and processing of spoken language. In the future, this kind of algorithms which imitate brain mechanisms could help machines to perceive the world around them.

Many people will have personal experience of how difficult it is for computers to deal with spoken language. For example, people who "communicate" with automated telephone systems now commonly used by many organisations need a great deal of patience. If you speak just a little too quickly or slowly, if your pronunciation isn't clear, or if there is background noise, the system often fails to work properly. The reason for this is that until now the computer programs that have been used rely on processes that are particularly sensitive to perturbations. When computers process language, they primarily attempt to recognise characteristic features in the frequencies of the voice in order to recognise words.

"It is likely that the brain uses a different process", says Stefan Kiebel from the Leipzig Max Planck Institute for Human Cognitive and Brain Sciences. The researcher presumes that the analysis of temporal sequences plays an important role in this. "Many perceptual stimuli in our environment could be described as temporal sequences." Music and spoken language, for example, are comprised of sequences of different length which are hierarchically ordered. According to the scientist's hypothesis, the brain classifies the various signals from the smallest, fast-changing components (e.g., single sound units like "e" or "u") up to big, slow-changing elements (e.g., the topic). The significance of the information at various temporal levels is probably much greater than previously thought for the processing of perceptual stimuli. "The brain permanently searches for temporal structure in the environment in order to deduce what will happen next", the scientist explains. In this way, the brain can, for example, often predict the next sound units based on the slow-changing information. Thus, if the topic of conversation is the hot summer, "su" will more likely be the beginning of the word "sun" than the word "supper".

To test this hypothesis, the researchers constructed a mathematical model which was designed to imitate, in a highly simplified manner, the neuronal processes which occur during the comprehension of speech. Neuronal processes were described by algorithms which processed speech at several temporal levels. The model succeeded in processing speech; it recognised individual speech sounds and syllables. In contrast to other artificial speech recognition devices, it was able to process sped-up speech sequences. Furthermore it had the brain's ability to "predict" the next speech sound. If a prediction turned out to be wrong because the researchers made an unfamiliar syllable out of the familiar sounds, the model was able to detect the error.

The "language" with which the model was tested was simplified it consisted of the four vowels a, e, i and o, which were combined to make "syllables" consisting of four sounds. "In the first instance we wanted to check whether our general assumption was right", Kiebel explains. With more time and effort, consonants, which are more difficult to differentiate from each other, could be included, and further hierarchical levels for words and sentences could be incorporated alongside individual sounds and syllables. Thus, the model could, in principle, be applied to natural language.

"The crucial point, from a neuroscientific perspective, is that the reactions of the model were similar to what would be observed in the human brain", Stefan Kiebel says. This indicates that the researchers' model could represent the processes in the brain. At the same time, the model provides new approaches for practical applications in the field of artificial speech recognition.


'/>"/>

Contact: Dr. Christina Schrder
cschroeder@cbs.mpg.de
49-034-199-40132
Max-Planck-Gesellschaft
Source:Eurekalert

Related biology news :

1. Invasion of the brain tumors
2. HIV is a double hit to the brain
3. AIDS interferes with stem cells in the brain
4. 60 second test could help early diagnosis of common brain diseases
5. U of MN researchers discover noninvasive diagnostic tool for brain diseases
6. U of Minnesota researchers discover noninvasive diagnostic tool for brain diseases
7. Influence of sex and handedness on brain is similar in capuchin monkeys and humans
8. Inside the brain of a crayfish
9. Specific brain protein required for nerve cell connections to form and function
10. Brains timing linked with timescales of the natural visual world
11. Adult brain can change, study confirms
Post Your Comments:
*Name:
*Comment:
*Email:
(Date:5/12/2016)... , May 12, 2016 WearablesResearch.com ... just published the overview results from the Q1 wave ... the recent wave was consumers, receptivity to a program ... data with a health insurance company. "We ... to share," says Michael LaColla , CEO of ...
(Date:5/3/2016)... 3, 2016  Neurotechnology, a provider of high-precision ... Automated Biometric Identification System (ABIS) , a complete ... MegaMatcher ABIS can process multiple complex biometric transactions ... of fingerprint, face or iris biometrics. It leverages ... and MegaMatcher Accelerator , which have been ...
(Date:4/28/2016)... -- First quarter 2016:   , Revenues ... first quarter of 2015 The gross margin was 49% ... and the operating margin was 40% (-13) Earnings per ... from operations was SEK 249.9 M (21.2) , Outlook ... 7,000-8,500 M. The operating margin for 2016 is estimated ...
Breaking Biology News(10 mins):
(Date:6/27/2016)... 27, 2016 /PRNewswire/ - BIOREM Inc. (TSX-V: BRM) ("Biorem" or ... its major shareholders, Clean Technology Fund I, LP and ... based venture capital funds which together hold ... a fully diluted, as converted basis), that they have ... entire equity holdings in Biorem to TUS Holdings Co. ...
(Date:6/27/2016)... ... 2016 , ... Parallel 6 , the leading software as a service ... Virtual Patient Encounter CONSULT module which enables both audio and video telemedicine communication ... , Using the CONSULT module, patients and physicians can schedule a face to face ...
(Date:6/27/2016)...  Liquid Biotech USA , ... Sponsored Research Agreement with The University of Pennsylvania ... cancer patients.  The funding will be used to ... clinical outcomes in cancer patients undergoing a variety ... employed to support the design of a therapeutic, ...
(Date:6/24/2016)... 24, 2016 Epic Sciences unveiled a ... susceptible to PARP inhibitors by targeting homologous recombination ... The new test has already been incorporated into ... cancer types. Over 230 clinical trials ... pathways, including PARP, ATM, ATR, DNA-PK and WEE-1. ...
Breaking Biology Technology: