iPhone 6s issue in recognising

[matteogobbi]

Dear,

We’re currently using OpenEars for a voice recognition exercise in our language learning app, busuu.

While we successfully implemented and tested on iPhone version =< 6, it looks like the microphone of the iPhone 6s has 50% more sensitivity.

As a result, when we start the voice recognition with the iPhone 6s, the OpenEars framework returns a result immediately as it has too much sensitivity.

As an example, when we start recording with an iPhone 6 and the threshold of noise of your SDK is already set at maximum, we get an average volume level of *-98/-100*, whereas on iPhone 6s it’s *-80/-85*

For your information, we cannot change the input gain of the iPhone 6s (Apple doesn’t allow this).

How could we solve this? Do you have any idea?

Looking forward to your answer,

Thanks

[Halle Winkler]

Welcome,

the threshold of noise of your SDK is already set at maximum

Which property does this line refer to?

[matteogobbi]

/**Speech/Silence threshhold setting. You may not need to make any changes to this, however, if you are experiencing quiet background noises triggering speech recognition, you can raise this to a value from 2-5 to 3.5 for the English acoustic model, and between 3.0 and 4.5 for the Spanish acoustic model. If you are experiencing too many words being ignored you can reduce this. The maximum value is 5.0 and the minimum is .5. For the English model, values less than 1.5 or more than 3.5 are likely to lead to poor results. For the Spanish model, higher values can be used. Please test any changes here carefully to see what effect they have on your user experience.*/

@property (nonatomic, assign) float vadThreshold;

I already set it at the max: 3.5 for the English acoustic model.

[Halle Winkler]

You can raise it as high as 4.4, does that help?

[matteogobbi]

I tried with 3.75 and looks ok.
Thanks.

[Halle Winkler]

Glad to hear it!

[Author]

Posts