Method to acquire preferred dynamic range function for speech enhancement
At least one exemplary embodiment is directed to a method of generating preferred dynamic range function to process audio reproduced by an earphone device. The function includes processing the audio to improve speech intelligibility. The function is acquired with a self-administered hearing test.
1. A method to generate a DRCF curve for a user comprising the steps of:
emitting a test signal into an ear canal of a user wearing an earphone;
measuring the test signal with a microphone;
evaluating the ear seal integrity of the earphone and sending a notice to the user if a seal value fails to exceed a threshold;
receiving an audio signal, referred to as the received audio signal;
generating a first dynamic range compression parameter set A, where the parameter set A includes at least one of a compression ratio value, an expansion ratio value, a threshold value, and a gate value;
generating a second dynamic range compression parameter set B, where the parameter set B includes at least one of a compression ratio value, an expansion ratio value, threshold value, and gate value;
processing the received audio signal with a first dynamic range compressor using the parameter set A to produce an output signal A;
processing the received audio signal with a second dynamic range compressor using the parameter set B to produce an output signal B;
selecting a preferred parameter set by selecting between the parameter set A and parameter set B by conducting a preference test by a user, where the user determines the preferred parameter set by comparing a speech intelligibility produced by using parameter set A and a speech intelligibility produced by using parameter set B, and generating a DRCF curve using the preferred parameter set.
2. The method according to claim 1 further including:
applying a gain to the received audio signal to generate a modified audio signal.
3. The method according to claim 2 , where the received audio signal is measured from an ambient sound microphone.
4. The method according to claim 3 where the modified audio signal is directed to an ear canal loudspeaker in an earphone.
5. The method according to claim 4 , where the received audio signal is at least one of speech audio and music audio.
6. The method according to claim 1 where the received audio signal is band pass filtered into multiple bands and each band is processed with a unique DRCF curve for each frequency band.
7. The method according to claim 1 , where the steps of claims are performed in an earphone.
8. The method of claim 7 , further including:
determining if the earphone used is correctly fitted.
9. The method of claim 8 , wherein the method to determine if the earphone used is correctly fitted comprises the steps of:
emitting a test signal into the earphone;
simultaneously cross-correlating an ear canal microphone signal with the emitted test signal;
comparing the result of the cross-correlation with a threshold correlation value to determine ear seal integrity; and
informing the user that the ear seal is not good if the cross-correlation value is significantly different from the threshold correlation value.