Hi all,
I'm currently working with VF3100.
I have experienced problems (echo issues) in some conditions: for example in case of poor-quality loudspeakers or if you change the speakers volume.
Among the vfctrl parameters that can be configured, I found NLATTENONOFF (disabled by default) and NLAEC_MODE.
I have not found any documentation about these parameters.
How does NL training mode work?
Can anyone help me use them?
Thanks,
Fabrizio
Non linear echo suppression on VF3100
-
- Member
- Posts: 15
- Joined: Fri Mar 02, 2018 6:05 pm
-
- Member++
- Posts: 22
- Joined: Fri Nov 07, 2014 9:03 am
Hello,
I m working with DSP features, too. Documentation is not clear for me:
NLATTENONOFF - Non-Linear echo attenuation
NLAEC_MODE - Non-Linear AEC training mode
By experiments:
param GAMMA_E and GAMMA_ETAIL controls echo response time, bigger number means faster response
param AECSILENCELEVEL , lowest number means bigger effect
Docs available:
xCORE VocalFusion Control Users Guide: https://www.xmos.com/developer/download ... .0rc7).pdf
xCore VocalFusion Fine Tuning Guide: https://www.xmos.com/developer/download ... rc5%29.pdf
VocalFusion:recommendations and considerations for testingand optimisingfar-field voice applications: https://www.xmos.com/developer/download ... 1.0%29.pdf
I m working with DSP features, too. Documentation is not clear for me:
NLATTENONOFF - Non-Linear echo attenuation
NLAEC_MODE - Non-Linear AEC training mode
By experiments:
param GAMMA_E and GAMMA_ETAIL controls echo response time, bigger number means faster response
param AECSILENCELEVEL , lowest number means bigger effect
Docs available:
xCORE VocalFusion Control Users Guide: https://www.xmos.com/developer/download ... .0rc7).pdf
xCore VocalFusion Fine Tuning Guide: https://www.xmos.com/developer/download ... rc5%29.pdf
VocalFusion:recommendations and considerations for testingand optimisingfar-field voice applications: https://www.xmos.com/developer/download ... 1.0%29.pdf
With kind regards
Tilde
Tilde