The first time and also the last time I was using a voice recognition software is around 10 years ago when I was working on my university thesis. It’s IBM ViaVoice. I was just tired of typing, but I had to give up after several days’ trying as I realized that it’s even more time consuming if you had to do the correction from time to time. iOS now comes with a simple voice recognition function too, but I have not seen anyone using it to play song, make call etc, and it does not work well with Chinese too.
I was totally amazed by Kouxun, the voice-to-text mobile application developed by iFlyteq (app store download). It’s deadly simple to use. Basically you just speak (Chinese mandarin only) with normal speed, the text will be auto typed in which can be sent as short message or email by one-click. Even the punctuations are auto inserted into the sentence. I’ve tried a couple of times, and the results are just perfect.
Kouxun requires the internet access as it’s actually a typical cloud-computing service. You voice is captured and its information is somehow sent to the cloud where it’s processed and return the text back to the phone.
Kouxun is probably the first app I’ve seen truly driven by technology over the cloud. Its owner iFlyteq was founded in June 1999 based on research originally conducted in the National Intelligent computer R & D Center and the Human-Machine Speech Communication Laboratory of the University of Science and Technology of China (USTC). The company’s principal products were a Chinese Text-to-speech engine and related software applications. “iFlyteq has over 10 years experience serving enterprise and government with the advanced voice recognition technology,” Jiang Tao, VP of iFlyteq said to me, “We saw the huge opportunity in current mobile industry where more and more consumers are using smart phone and get used to internet access over mobile device”.
“Our mission is simple”, Jiang said, “Set your Fingers Free.” Nobody knows if Apple plans to enhance its voice recognition function in future version of iOS. I hope it does as sometime voice is just much more convenient and efficient than moving fingers.
Apple is planning to incorporate voice…with its partnership with Nuance. The performance for Chinese recognition is comparable to Kouxun and I’d say better than Google.
Thanks for the link, Alan. It seems that we just look at the right market… 😉
Leave a comment