Capturing Cat's sound in normal environment
With high density spectral data of sound, we can expand the possible usage of the analysis engine. Other than getting extra detail, another advantage is the separation power of components. In another word, it boosts the Signal to Noise Ratio in analysis.
One of the uses is the collection of cat’s sound. If you look at the nature of cat’s sounds, they last over 0.5s, and no rapid changes like human pronunciation. So it shouldn’t have any problem if we try to analyze them. However, the sound energy of cats’ sounds is relatively low with a normal background environment. This makes the collections of voice samples difficult and hard to put into any practical use.?
Now, with our analysis data, we can do recognition of cat’s sound.?
Here is link to the 3D object file with the sound sample, which you can view on any 3D object viewer
And the source sound file?
Capturing a cat’s meow sound in a daily environment is always with background noise. With our technology those can be separated. Like the sample above, if viewing with conventional spectral data(in GREEN). The noise presents in all the spectrum, but after our isolation with only muscular activities, only human voice and cat’s sound remains(in RED, That’s our another business secret).? This enables any further study, measuring or recognition on those cat sounds.?
There are some regions that cross over in human voice and cat’s sound, so we would need further study on muscular behavior models to separate them.?
This tool opens the door not only in human pronunciation, but also animal sounds which are troubled by the noisy environments.?
@sound analysis
@cat
@innovation