Minimum system
The minimum system of a speech recognition system in a general application needs to include: a controller module, a speech recognition module, an amplifier module, and a speaker. If you need to use the voice recognition function, you need to add a microphone in the system.
As shown in FIG. 2 , when the user uses the voice recognition or the voice wake-up function, the host computer sends a command to start a voice recognition or a voice wake-up function to the voice module, and the module converts the voice data collected from the microphone through an internal identification module. The corresponding recognition result is returned to the controller through the communication interface.
Figure 2 - System configuration diagram using Syn7318 speech recognition module
Application
With the development of technology, traditional keyboards and buttons have not been able to meet the needs of people, and more convenient, more natural and humane human-computer interaction has become the direction that people are pursuing. Hearing-based human-computer interaction is undoubtedly one of the most interactive friendly ways. Because of its simple structure and convenient construction, embedded speech interaction has become a hot topic in current research. Compared with the PC speech recognition system, the embedded speech recognition system has the advantages of small size, low power consumption, high reliability, low investment, flexible installation, etc. In the smart home, robotics and consumer electronics and other fields.