The Institute of Scientific and Industrial Research, Osaka University

contact
home
japanese
HOME > Activity Reports > Research Activities

Research Activities

AI and Folk Duo WARAINAKI have composed a Red Feather campaign song.

 We propose a method to locate relations and constraints between a music score and its impressions,by which we show that AI (Artificial Intelligence) and machine learning techniques may provide a powerful tool for composing music and analyzing human feelings.
 We examine its generality by modifying some arrangements and creating some compositions to provide the subjects with a specified impression.
 This introduces some user interfaces, which are capable of predicting feelings and creating new objects based on seed structures, such as spectrums and their transition for sounds that have been extracted and are perceived as favorable by the user.
 This AI collaborates with Folk Duo WARAINAKI, and composes a Red Feather campaign song in Nara.

  • Figure 1
    Fig1.Automatic composition system that was developed

  • Figure 1
    Fig2.Score of the music that was created (in part)

Related Link
Numao Laboratory, Department of Architecture for Intelligence
http://www.ai.sanken.osaka-u.ac.jp/

Keywords:
Engineering, AI(Artificial Intelligence), Machine Learning, Evolutionary Computation, Automatic Composition

Top of Page

Diverse Adsorption/Desorption Abilities Originating from the Nanostructural Morphology of VOC Gas Sensing Devices Based on Molybdenum Trioxide Nanorod Arrays

3D-network of ultra-fine single-crystal α-MoO3 nanorod arrays based gas sensors show a prompt response and discrimination to VOCs. S. Cong, T. Sugahara, and co-workers illustrate the performance of the sensors strongly depend on the specific morphologies of the nanorod arrays, such as length, number and coverage of nanorods in the 3D network. The arrays are spontaneously grown by a simple single-step solution route. A prompt response and obvious discrimination of ethanol, methanol, isopropanol and acetone vapors at 573 K are investigated via the modulation of the resistance of the gas sensors. The conductance modulation of the nanorod arrays are attributed to the hydrogen ions decomposed from VOCs intercalated into the van de Waals’s gaps of layered α-MoO3 and the subsequent reduction the cornered oxygen to H2O. The sensitivity, response time and recovery time of the sensors strongly depend on the specific morphologies of the nanorod arrays, such as length, number and coverage of nanorods in the 3D network. A reaction mechanism in which the 3D-network nanorod arrays adsorb and react with the target molecules more readily than the seed layer is proposed to explain the different response and recovery times of the sensors. These random 3D-network nanorod arrays with functionally tunable morphology are promising for universal application as gas sensors for detecting various vapors, and provide valuable insights for the production of fast, large-scale, low-cost and simple synthesis of sensing devices.

  • Figure 1
    Figure. a Photo figure of the gas sensor device, and b Cross-sectional FE-SEM image of
    the MoO3 nanorod arrays.

This work has been published online by Advanced Materials Interfaces on May 27, 2016. (see http://onlinelibrary.wiley.com/doi/10.1002/admi.201600252/abstract) and will be appeared on the inside cover of the July 22, 2016 issue.

Keywords:
oxide semiconductors, MOD (Metal Organic Decomposition) Method, gas sensors, VOCs (Volatile Organic Compounds)

Top of Page

Prof. Takashi Washio received 2016 IBM Faculty Award from IBM

Prof. Takashi Washio in the Institute of Scientific and Industrial Research, Osaka University received 2016 IBM Faculty Award from IBM (International Business Machines Corporation). IBM Faculty Award is an award for a full professor having outstanding contribution to a strategic and challenging research field and selected from many candidate professors in research oriented universities all over the world. This award aims to encourage the research of the professor and promote further development of the research field.
Prof. Takashi Washio was awarded upon his past research outcomes in machine learning and data mining and also for his future studies on application of machine learning to advanced sensing technology and development of social services using the technology in the IoT era. His winning memorial lecture will be given in HICSS-50: Hawaii International Conference on System Sciences to be held on Hawaii island on 4th-7th, January, 2017.

Top of Page

back number
2016
2015
2014
2013

The Institute of Scientific and Industrial Research, Osaka University

contact home japanese
HOME > Activity Reports > Research Activities

Research Activities

AI and Folk Duo WARAINAKI have composed a Red Feather campaign song.

 We propose a method to locate relations and constraints between a music score and its impressions,by which we show that AI (Artificial Intelligence) and machine learning techniques may provide a powerful tool for composing music and analyzing human feelings.
 We examine its generality by modifying some arrangements and creating some compositions to provide the subjects with a specified impression.
 This introduces some user interfaces, which are capable of predicting feelings and creating new objects based on seed structures, such as spectrums and their transition for sounds that have been extracted and are perceived as favorable by the user.
 This AI collaborates with Folk Duo WARAINAKI, and composes a Red Feather campaign song in Nara.

  • Figure 1
    Fig1.Automatic composition system that was developed

  • Figure 1
    Fig2.Score of the music that was created (in part)

Related Link
Numao Laboratory, Department of Architecture for Intelligence
http://www.ai.sanken.osaka-u.ac.jp/

Keywords:
Engineering, AI(Artificial Intelligence), Machine Learning, Evolutionary Computation, Automatic Composition

Top of Page

Diverse Adsorption/Desorption Abilities Originating from the Nanostructural Morphology of VOC Gas Sensing Devices Based on Molybdenum Trioxide Nanorod Arrays

3D-network of ultra-fine single-crystal α-MoO3 nanorod arrays based gas sensors show a prompt response and discrimination to VOCs. S. Cong, T. Sugahara, and co-workers illustrate the performance of the sensors strongly depend on the specific morphologies of the nanorod arrays, such as length, number and coverage of nanorods in the 3D network. The arrays are spontaneously grown by a simple single-step solution route. A prompt response and obvious discrimination of ethanol, methanol, isopropanol and acetone vapors at 573 K are investigated via the modulation of the resistance of the gas sensors. The conductance modulation of the nanorod arrays are attributed to the hydrogen ions decomposed from VOCs intercalated into the van de Waals’s gaps of layered α-MoO3 and the subsequent reduction the cornered oxygen to H2O. The sensitivity, response time and recovery time of the sensors strongly depend on the specific morphologies of the nanorod arrays, such as length, number and coverage of nanorods in the 3D network. A reaction mechanism in which the 3D-network nanorod arrays adsorb and react with the target molecules more readily than the seed layer is proposed to explain the different response and recovery times of the sensors. These random 3D-network nanorod arrays with functionally tunable morphology are promising for universal application as gas sensors for detecting various vapors, and provide valuable insights for the production of fast, large-scale, low-cost and simple synthesis of sensing devices.

  • Figure 1
    Figure. a Photo figure of the gas sensor device, and b Cross-sectional FE-SEM image of
    the MoO3 nanorod arrays.

This work has been published online by Advanced Materials Interfaces on May 27, 2016. (see http://onlinelibrary.wiley.com/doi/10.1002/admi.201600252/abstract) and will be appeared on the inside cover of the July 22, 2016 issue.

Keywords:
oxide semiconductors, MOD (Metal Organic Decomposition) Method, gas sensors, VOCs (Volatile Organic Compounds)

Top of Page

Prof. Takashi Washio received 2016 IBM Faculty Award from IBM

Prof. Takashi Washio in the Institute of Scientific and Industrial Research, Osaka University received 2016 IBM Faculty Award from IBM (International Business Machines Corporation). IBM Faculty Award is an award for a full professor having outstanding contribution to a strategic and challenging research field and selected from many candidate professors in research oriented universities all over the world. This award aims to encourage the research of the professor and promote further development of the research field.
Prof. Takashi Washio was awarded upon his past research outcomes in machine learning and data mining and also for his future studies on application of machine learning to advanced sensing technology and development of social services using the technology in the IoT era. His winning memorial lecture will be given in HICSS-50: Hawaii International Conference on System Sciences to be held on Hawaii island on 4th-7th, January, 2017.

Top of Page

back number
2016
2015
2014
2013