 |
Unique, the World Leader
This technology breakthrough originates from the neural network speech analysis technology. It is continualy improved through collective research from UCLA, UC Berkeley, Texas Instruments, Tsinghua University.
The technology is based on Psychoacoustics Modeling, which analyzes human voice and establishes the score models. By adopting discriminative modeling methods of statistics models, the whole system can be made compact, noise robust and accurate, maintaining consistency with human perception habits. The edge of this technique is its good performance under most environmental noise, where all the other technologies will fail.
There is no other company, except Arivoc, can provide such an intelligent oral training technology. |
Other speech technology includes:
The pronunciation analysis and correcting technology, engines (PC edition and embedded system) —— Speech recognition technology and engines (PC edition and embedded) —— Speech synthesis technology and engines —— Human-machine conversation technology —— Speaker recognition technology
Our pronunciation analysis technology is the best in the world. Please try Product Demo
Our speech recognition technology is among the top 4 in the world, which is widely used in speech control (car, home appliance) and the man-machine dialogue system.
Other technologies are also of first class in the world.
The embedded engine is suitable for the development on ARM chips.
|
|
