DescriptionBack to top
- Multi-language speech recognition module with serial interface
- Supports up to 28 custom Speaker Independent (SI) command vocabularies
- Supports up to 32 user-defined Speaker Dependent (SD) triggers or commands
- SonicNET technology for wireless communications
The EasyVR Speech Recognition Module 2.0 is a multi-purpose speech recognition module designed to easily add versatile, robust and cost effective multi-language speech recognition capabilities to almost any application. The module includes a host of built in Speaker Independent commands for ready-to-run basic controls. It supports up to 32 user-defined Speaker Dependent (SD) triggers or commands as well as voice passwords. Speaker Dependent commands can be spoken in any language. The EasyVR 2.0 also introduces the ability for users to create up to 28 of their own Speaker Independent (SI) Command Vocabularies using the QuickT2SI Lite Software (license sold separately). Additionally the EasyVR 2.0 includes SonicNet technology for wireless communication between modules or any other sound source. DTMF tone generation is also included.
What's IncludedBack to top
- EasyVR 2.0 module
- Software (free download)
MultimediaBack to top
Supplier Product CodeBack to top
Write Your Own Review
Easy to use, reasonalby accurate. Best deal for the money!Review by RonBack to top
I got the easyVR board and immediately discovered that I needed the development board to really use the board effectively. No problem, another 15 bucks. It took a while to figure out the software but you can do it in an hour or two and have an arduino uno recognizing speech words.
The main shortcoming of the system is background noise, which is a problem with all speech recognizers. The system would not recognize while I had the window air conditioner running in the room. With the room dead silent the unit is very good. The unit has AGC that will take whatever background noise there is and put it to a reference level. So even fairly low volume background noise will interfere with the recognition.
A second problem seems to be the AGC adjustment to the voice input. I found that the unit would not recognize after ten minutes or so. After I spoke several words it began recognizing and was then accurate. Seems to take a bit of speech for it to get the volume level correct.
The last problem I found is that there are some words you cannot program. It refuses to accept words in its internal vocabulary for some reason. This can be a real hassle if you need only a number or two and not the whole range of numbers. Using the entire built-in library of numbers of course uses more of your precious memory! I guess you could go into the source code and chop out the stuff you don't want.
The supplied software generator is very clever! It will generate a compilable source code with your speech words for the arduino uno for sure, and possibly others. The code generated is heavy so you will want to chop out a lot of it in your final application to save memory, but it gives you an instant runnable program to put in your project. That is really a huge advantage for new programmers! (and for us old hats too!)
The only complaint I have with the package is the documentation, which is sort of haphazard. If you follow exactly what they show you will get results, but there are a lot of questions you will need to spend time figuring out for yourself.
(Posted on 9/3/14)