Difference between revisions of "Orange Pi AI Stick Lite"
(→Notes) |
(→Notes) |
||
(One intermediate revision by the same user not shown) | |||
Line 46: | Line 46: | ||
== '''Notes''' == | == '''Notes''' == | ||
<br> | <br> | ||
− | + | 1. Compatibility: It is workable with H2, H3, H5, H6, A64 Orange Pi development boards.<br> | |
<br> | <br> | ||
− | + | 2. Model transformation tools will be provided which supports model decomposition of convolutional neural network (CNN) based on caffe.<br> | |
<br> | <br> | ||
− | + | 3. PLAI training tool can directly apply the original image, video, voice, natural language rapid training prototyping to the neural network computing stick.<br> | |
<br> | <br> | ||
== '''Product parameter(Refer to the official website details page)''' == | == '''Product parameter(Refer to the official website details page)''' == |
Latest revision as of 17:51, 22 June 2022
Contents
Orange Pi AI Stick Lite Introduction
Orange Pi AI Stick Lite has a built-in Lightspeeur SPR2801 AI chip, which empowers mobile phones, PCs and other devices to run real-time deep learning and achieve picture, video, speech and natural language recognition directly through a USB interface.
Features
- Supports USB 2.0 and 3.0 standard interface communication
- No programming required, no language barrier
- Open SDK for X86, ARM and other platforms
- Support for Android, Linux and other operating systems
- Support for neural network models such as VGG and SSD
What the Orange Pi AI Stick Lite can do?
- Target detection & classification
- Mobile edge computing
- Intelligent surveillance
- Smart toys and robots
- Smart homes
- Virtual reality and augmented reality
- Face detection and recognition
- Speech recognition
- Natural language processing
- Embedded deep learning devices
- Cloud-based machine learning and deep learning systems
- Artificial intelligence data centre servers
- Advanced assisted driving and automated driving
Notes
1. Compatibility: It is workable with H2, H3, H5, H6, A64 Orange Pi development boards.
2. Model transformation tools will be provided which supports model decomposition of convolutional neural network (CNN) based on caffe.
3. PLAI training tool can directly apply the original image, video, voice, natural language rapid training prototyping to the neural network computing stick.