A General Architecture for Self-Adaptive AmI Components Applied in Speech Recognition

Size: px

Start display at page:

Download "A General Architecture for Self-Adaptive AmI Components Applied in Speech Recognition"

Philip Dean
5 years ago
Views:

1 AG Softwarearchitecture A General Architecture for Self-Adaptive AmI Components Applied in Speech Recognition Holger Klus ICSE AG Softwarearchitektur 1

2 Outline Outline Application Scenario Our three types of adaptation Integration of speech recognition into our AmIscenario as one type of adaptation Outlook AG Softwarearchitektur 2

3 Motivation Example Assisted Training (1) Training group of (racing) cyclists Given track profile Track Section Uphill Even Downhill Cyclists have different qualities regarding the track sections Each cyclist has a given overall physical condition and an individual training plan Training effect depends on the speed of the group and the position of the cyclist within the group formation Main goal of the system Optimization of the single training effect Optimization of the group training effect AG Softwarearchitektur 3

4 Motivation Example Assisted Training (2) GPS Bluetooth / WLAN AmI PDA GSM pulse sensor trainer wind sensor MICA Nodes Internet-based training database pedal power sensor AG Softwarearchitektur 4

Motivation Example Assisted Training (3) (Currently) explored adaptation variants Service Implementation Adaptation Self-adaptation of speech recognition service Service Usage Adaptation show pulse

5 Motivation Example Assisted Training (3) (Currently) explored adaptation variants Service Implementation Adaptation Self-adaptation of speech recognition service Service Usage Adaptation show pulse and warning (text) Change from a visual UI to a speech generation UI service implementation Service Configuration Adaptation Headphone LCD Display Mobile Phone Display New component containing new services plugs into the training system and results in a reconfiguration of the overall system from single bicycle to group bicycle training Pulse group training monitor pulse set limits Group Control Control Application Application Pulse Sensor get current pulse AG Softwarearchitektur 5

6 Service Implementation Adaptation Speech Recognition Speech Recognition in the AmI-Prototype Why» Natural communication with the system» But important: High recognition rate in changing environmental conditions such as Different speaker (young, old, female, male) Background noise Mental and physical stress of the speaker Our goal» Input of upper and lower pulse boundary using speech input» Take into account the physical stress of the user Realization in cooperation with the BUTE (Budapest University of Technology and Economics)» Using Hungarian speech recognition engine Task: Mapping on German speech» Take into account current stress of the speaker using German speech recordings» Feature: Speech recognizer is runnable on small devices like PDA, Handy, AG Softwarearchitektur 6

7 Speech Recognition Reference Model Speech Signal Matching Algorithm Recognized Phrase AG Softwarearchitektur 7

8 Speech Recognition under Physical Stress of the Speaker Generic Approach Stress Recognition» Current research topic Automatic stress detection» In our scenario Stress detection using the pulse value of the speaker Compensation of stress features» Manipulation of speech-statistics at runtime High computational power required» Switching between different speech-statistics which have been created offline AG Softwarearchitektur 8

9 Architecture of a Self-Adaptive Component Pulse Sensor Sensor Network ContextProvidingInterface Self-adaptive speech recognizer Stress recognition and stress level Self-adaptive AmI component Context Providing Switching between statistics Context Adaptation Context Processing Decision of how to react on a specific stress level ContextAdaptationInterface Actor Network AG Softwarearchitektur 9

10 Outlook Service configuration adaptation First version is already implemented Description techniques for Quality of Service Especially required by» Service Usage adaptation» Service Configuration adaptation What is the best service at the moment? Criterion can change during runtime! AG Softwarearchitektur 10

11 Questions and Discussion Any Questions? AG Softwarearchitektur 11

Radio Frequency Integrated Circuits Prof. Cameron Charles

Radio Frequency Integrated Circuits Prof. Cameron Charles Overview Introduction to RFICs Utah RFIC Lab Research Projects Low-power radios for Wireless Sensing Ultra-Wideband radios for Bio-telemetry Cameron