Jun 30, 2021

Q-life Technology Midas Touch Case Series 008 When The Offline Speech Recognition Control Module Meets The Air Fan PART I

Leave a message

1

As soon as the May Day holiday is over, the country will enter the "barbecue" mode of summer. As a common home appliance used to cool down the summer heat, electric fans have filled major shopping malls and are ready to enter thousands of households to make a contribution to people's summer life.

Fan control mode has been developing with the progress of science and technology constantly, from the rope control to machinery knob control then touch control, infrared remote, intelligent mobile phone APP control, intelligent speaker speech recognition switch control, to offline speech recognition control and so on, the various kinds of control mode people’s life more and more convenient, user experience is becoming much better than before。

The speech recognition controlled fan has been sold in the market several years ago which was present by the intelligent speaker manufacture, but from the feedback of the market, the sales situation are not ideal. Why? Because the using experience for the users is not good. Why is the experience bad? Because it use the online speech recognition control scheme. Why does online speech control feel bad? Let's do a quick analysis.

Online speech recognition control fans generally uses the online speech recognition function of smart speakers to recognition man-machine dialogue, and converts the commands recognized by the cloud voice search engine into digital control commands. When the voice reply is played, the digital control code is sent out through RF communication. The fan with RF receiving module executes the control command after receiving the digital control code.

This kind of speech recognition needs a complicate process to realize. The intelligent speakers picked your voice commands and use Wi-Fi networks to upload this voice signal to the cloud voice search engine, and then return to the recognition result, this process will be affect by the smart speaker usage scenarios such as Wi-Fi signal strength and network speed, and the performance of the cloud speech recognition efficiency of search engine. The Wi-Fi signal strength is one of the biggest factors. In the household environment the uneven Wi-Fi signal strength is very common. Some corner almost no Wi-Fi signals are also common situation. Therefore, from your launch the voice commands, to through the layers of transformation, converts voice commands into digital control signal to control the fan, and get feedback by the speakers, A time delay of 2 to 5 seconds is normal, and sometimes it is possible to be completely unusable due to a dead Wi-Fi signal or off connection. The poor experience of a delayed response to voice commands, or even no response at all, makes users feel that the product is "not smart enough."


Send Inquiry