Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
Clone this wiki locally
This page outlines the core functions of the µSpeech library.
At its heart the µSpeech library is made of a single
signal object. The signal object contains both speech recognition APIs and some basic DSP APIs.
The Signal API provides a raw interface to the microphone signal using nicely wrapped C++ objects. Methods not listed here are deprecated and will be removed.
The Signal API
Is the constructor for the signal object.
This line calibrates the voice object so as to remove noise and interference from the system.
This returns the closest phoneme guess to the real phoneme, if it is too quiet the system will return
' ' and
'm' if there was an error. Read more...
General Utility methods
int signal::snr(int power);
Returns the signal to noise ratio of a filter.
Returns total power, average power can be derived by simply dividing the result by 32.
The maximum power detected by the microphone.
Contains the raw microphone output.
unsigned int signal.maxPower;
For debugging purposes.