Hi,
I’ve been implementing a “full hearing aid software” based on the WDRC_8BandFIR_Stereo example from the Tympan examples, and I’ve added noise reduction, which works great. But lately I’ve been experiencing a very loud high pitched noise mostly whenever there’s silence/no one speaking, if I have noise reduction off, independent from compression (doesn’t matter if it’s on or off). This happen if I have a global digital gain of like 5.0 dB and the following parameters of compression:
BTNRH_WDRC::CHA_DSL dsl_left = {5, // attack (ms)
50, // release (ms)
115, // maxdB. calibration. dB SPL for input signal at 0 dBFS. Needs to be tailored to mic, spkrs, and mic gain. (original = 115)
0, // 0=left, 1=right…ignored
N_CHAN, // num channels
{317.1666, 502.9734, 797.6319, 1264.9, 2005.9, 3181.1, 5044.7}, // cross frequencies (Hz)
{0.7, 0.7, 0.7, 0.7, 0.7, 0.7, 0.7, 0.7}, // compression ratio for low-SPL region (ie, the expander…values should be < 1.0)
{30.0, 30.0, 30.0, 30.0, 30.0, 30.0, 30.0, 30.0}, // expansion-end kneepoint
//{55, 55, 65, 65, 85, 85, 85, 85}, // Audiogram values
{21, 21, 29, 29, 45, 45, 45, 45}, // compression-start gain
{0.7f, 0.9f, 1.0f, 1.1f, 1.2f, 1.4f, 1.5f, 1.5f},
{30.0, 30.0, 30.0, 30.0, 30.0, 30.0, 30.0, 30.0}, // compression-start kneepoint (input dB SPL)
{79.f, 88.f, 91.f, 93.f, 98.f, 103.f, 102.f, 100.f} // output limiting threshold (comp ratio 10)
};
When I say I turn off compression, I just load up another preset with the CR’s set to 1.0, but with the same gain values as the preset above.
Any tips?
Thanks!