4:05 PM
danyrolux
AI Voice assistant has become the mainstream technology of smart home. Rockchip launched AI voice assistant solution RK3229 and exhibit two types of smart speakers at CES 2017, highlighting its features:
- Supports digital and analog ring mic arrays and line-mic arrays
- Supports third-party Chinese and English semantic recognition platform
- Supports multi-microphone array algorithm，e.g：Sound source localization / Echo cancellation / De-reverberation / Arbitrary beam forming / Noise reduction / Far-field recognition / Voice wakeup.
Rockchip RK3229 is most significant solution for AI market. Quad-core Cortex-A7 CPU, up to 1.3GHz; Mali-400MP2 GPU, support OpenGL ES1.1/2.0; Embedded Audio DAC;Dual I2S interface, support I2S and PCM; Supports 8 channel I2S RX/TX; Embed RGMII interface and 100M MAC PHY.
AI intelligent platform based on Rockchip RK3229 can realize voice control, inputting voice through multi mics to achieve better noise reduction. It can improve the long-distance voice recognition performance, and greatly raise the accuracy of man-machine interaction.
In the future, voice assistant can connect with all AI intelligent terminals at home. You may control TV, smart appliances and any other IoT devices at home via voice assistant, also assisting customers to achieve intelligent terminal solution and building “Home AI”. voice assistant can be applied to different business partners’ platform, such as taxi service, taking-away service, and music video streaming media services, as well as financial services, etc.
Rockchip has long ago been engaging in AI industry and already launched many solutions including face recognition, voice control VR and accompanying robot.
As Rockchip’s flagship chip, RK3399 adopts a big.LITTLE structure (dual-core Cortex-A72+quad-core Cortex-A53) CPU and ARM Mali-T860MP4 GPU, providing high performance, low power consumption, and with outstanding media performance and extendibility. It supports HDMI2.0 port and H.264/H.265/VP9 4K 10bit@60fps video decoding, and the PCI-e based high speed Wi-Fi and storage. The face recognitions tablet PC equipping with RK3399 has the following three features:
1. Support offline application
2. High accuracy and threshold adjustable according to different scenes
3. High speed and low power consumption
Rockchip RK3288 conforming to second standard of VR all-in-one adopts Quad-core Cortex-A17 CPU and Mali-T764 GPU, allowing latency lower than 20ms, refreshing’ rate higher than 75Hz, and a gyroscope refresh rate higher than 1K. It supports optical software against distortion and dispersion, as well as pupil distance regulating algorithm, to effectively solve the problem of dizziness from the aspects of definition, visual distortion processing, gyroscope precision and stability.
RK3288 man-machine interaction accompanying robot solution is equipped with memory reduction technology, real-time light engine game technology and many other advanced technologies. It also supports man-machine cooperation, voice interaction and remote control functions.
- Supports digital and analog ring mic arrays and line-mic arrays
- Supports third-party Chinese and English semantic recognition platform
- Supports multi-microphone array algorithm，e.g：Sound source localization / Echo cancellation / De-reverberation / Arbitrary beam forming / Noise reduction / Far-field recognition / Voice wakeup.
Rockchip RK3229 is most significant solution for AI market. Quad-core Cortex-A7 CPU, up to 1.3GHz; Mali-400MP2 GPU, support OpenGL ES1.1/2.0; Embedded Audio DAC;Dual I2S interface, support I2S and PCM; Supports 8 channel I2S RX/TX; Embed RGMII interface and 100M MAC PHY.
AI intelligent platform based on Rockchip RK3229 can realize voice control, inputting voice through multi mics to achieve better noise reduction. It can improve the long-distance voice recognition performance, and greatly raise the accuracy of man-machine interaction.
In the future, voice assistant can connect with all AI intelligent terminals at home. You may control TV, smart appliances and any other IoT devices at home via voice assistant, also assisting customers to achieve intelligent terminal solution and building “Home AI”. voice assistant can be applied to different business partners’ platform, such as taxi service, taking-away service, and music video streaming media services, as well as financial services, etc.
Rockchip has long ago been engaging in AI industry and already launched many solutions including face recognition, voice control VR and accompanying robot.
As Rockchip’s flagship chip, RK3399 adopts a big.LITTLE structure (dual-core Cortex-A72+quad-core Cortex-A53) CPU and ARM Mali-T860MP4 GPU, providing high performance, low power consumption, and with outstanding media performance and extendibility. It supports HDMI2.0 port and H.264/H.265/VP9 4K 10bit@60fps video decoding, and the PCI-e based high speed Wi-Fi and storage. The face recognitions tablet PC equipping with RK3399 has the following three features:
1. Support offline application
2. High accuracy and threshold adjustable according to different scenes
3. High speed and low power consumption
Rockchip RK3288 conforming to second standard of VR all-in-one adopts Quad-core Cortex-A17 CPU and Mali-T764 GPU, allowing latency lower than 20ms, refreshing’ rate higher than 75Hz, and a gyroscope refresh rate higher than 1K. It supports optical software against distortion and dispersion, as well as pupil distance regulating algorithm, to effectively solve the problem of dizziness from the aspects of definition, visual distortion processing, gyroscope precision and stability.
RK3288 man-machine interaction accompanying robot solution is equipped with memory reduction technology, real-time light engine game technology and many other advanced technologies. It also supports man-machine cooperation, voice interaction and remote control functions.
Don’t miss any of our future video tutorials, follow us on Youtube. Like us on Facebook. Add us in your circles on Google+. Watch our photo albums on Flickr. Subscribe now to our newsletter.
0 comentarii:
Post a Comment