Several little bugs with latest version

[ad2003]

Hi! Thanks for the updates — amazing program!

I’m having a couple of issues with version 3.6.2 while using VB-Audio Cable / HiFi Cable. I’m not sure whether these are caused by my setup or by ASH, so I wanted to report them.

File deletion errors when using Quick Configurations

Whenever I run a quick configuration, I repeatedly get PermissionError and WinError 32 messages during the BRIR export phase. It looks like ASH is unable to delete existing files in:

C:\Program Files\EqualizerAPO\config\ASH-Outputs\BRIRs\live_dataset

Here is the relevant log output:

2025-11-22 21:25:55 - root - INFO - Loading reverberation data
2025-11-22 21:25:55 - root - INFO - Loading HRTF data
2025-11-22 21:25:55 - root - INFO - Adjusting levels
2025-11-22 21:25:55 - root - INFO - Preparing reverberation data
2025-11-22 21:25:55 - root - INFO - Integrating HRIRs and reverberation
2025-11-22 21:25:56 - root - INFO - Processing low frequencies
2025-11-22 21:25:57 - root - INFO - Applying EQ
2025-11-22 21:25:59 - root - INFO - Finalising dataset
2025-11-22 21:25:59 - root - INFO - Execution time:3.9835011959075928 seconds
2025-11-22 21:25:59 - root - INFO - Preparing BRIRs for export
2025-11-22 21:26:00 - root - INFO - Warning: Could not delete file 'C:\Program Files\EqualizerAPO\config\ASH-Outputs\BRIRs\live_dataset\backup': [WinError 5] Access is denied: 'C:\Program Files\EqualizerAPO\config\ASH-Outputs\BRIRs\live_dataset\backup'
Traceback (most recent call last):
File "ash_toolset\brir_export.py", line 194, in export_brir
PermissionError: [WinError 5] Access is denied: 'C:\Program Files\EqualizerAPO\config\ASH-Outputs\BRIRs\live_dataset\backup'
2025-11-22 21:26:00 - root - INFO - Warning: Could not delete file 'C:\Program Files\EqualizerAPO\config\ASH-Outputs\BRIRs\live_dataset\BRIR_E0_A-135.wav': [WinError 32] The process cannot access the file because it is being used by another process
Traceback (most recent call last):
File "ash_toolset\brir_export.py", line 194, in export_brir
PermissionError: [WinError 32] The process cannot access the file because it is being used by another process
... (and similar repeated messages for A-40, A-90, A0, A135, A40, A90)
2025-11-22 21:26:00 - root - INFO - Cleared directory: C:\Program Files\EqualizerAPO\config\ASH-Outputs\BRIRs\live_dataset (deleted=7, failed=8)
2025-11-22 21:26:00 - root - INFO - BRIR WAV set saved to: C:\Program Files\EqualizerAPO\config\ASH-Outputs\BRIRs\live_dataset
2025-11-22 21:26:00 - root - INFO - Execution time:0.5570003986358643 seconds
2025-11-22 21:26:03 - root - INFO - HpCF (WAV FIR): KZ ZSTX Sample A saved to: C:\Program Files\EqualizerAPO\config\ASH-Outputs\HpCFs\FIRs\KZ
2025-11-22 21:26:03 - root - INFO - Applied filter for KZ / KZ ZSTX / Sample A

It looks like either ASH, EqualizerAPO, or the OS is locking these files. This didn’t happen for me in previous versions. Or at least, I never saw it.

Sample-rate mismatch with VB Cable / HiFi Cable

I recently changed my audio setup and adjusted some device settings.

I’m using VB-Audio Cable or HiFi Cable set to 96 kHz, and my main sound device is also set to 96 kHz. However, ASH reports a sample-rate mismatch.

I also tested 48 kHz and 44.1 kHz:

At 44.1 kHz, ASH shows no mismatch.

When creating a Binaural Room Setup, I have to manually set ASH to 48 kHz or it won’t run.

For Headphone Correction, I need to set ASH to 44.1 kHz, otherwise it won’t run.

If I set everything to 48 kHz, ASH works, but still shows a mismatch with my device.

This behavior did not occur before — previously I could keep everything at one rate (e.g., 48 kHz).

(1) VB Cable configured at 96 kHz
(2) ASH showing 44.1 kHz mismatch warning even when devices are set higher

ASH resets my selected Headphone Correction profile

When I open ASH, it automatically selects the first HpCF profile in the list and overwrites my previous selection. I don’t remember this happening before version 3.6.2.

Summary

BRIR export fails to delete or overwrite some files (WinError 5 / WinError 32).

ASH detects incorrect sample rate even when devices are configured correctly.

BRS requires 48 kHz, HpCF requires 44.1 kHz, otherwise they fail to run.

ASH resets my selected HpCF profile on startup.

If you need any additional logs, configs or testing from my end, I’m happy to help. The tool is amazing and I appreciate all the work that goes into it!

[ad2003]

for deleting the files I ran net stop audiosrv net start audiosrv this worked out.

I also put everything to 44.1khz now and it seems to work. Are there any advantages using smple rates higher than 44.1khz?

thx

[ShanonPearce]

BRIR export fails to delete or overwrite some files (WinError 5 / WinError 32).

I can confirm the file write error "PermissionError: [WinError 5] Access is denied:" can safely be ignored. The step to clear out previous files in the output folder can fail if some of the files are locked by equalizer apo (or some other process). Its a clean up step that removes excess files and shouldnt cause any problems in this scenario. The step that follows is it will write the new wav files which successfully overwrites the data in the existing wav files so the data will still be updated. I might change the logging a bit to be clearer.

ASH detects incorrect sample rate even when devices are configured correctly.

Would you be able to confirm the following:

• Default audio device (in windows sound control panel)
• Default audio device sample rate (in windows sound control panel)
• Default audio device (in ASH quick config tab -> audio device format)
• Default audio device sample rate (in ASH quick config tab -> audio device format)
• Selected sample rate (in ASH quick config tab -> IR format)
• Are filters/parameters applied again after changing sample rate?
• If you open equalizer APO configuration editor then open details for ASH-Outputs\E-APO-Configs\ASH_Toolset_Config.txt (green arrow), does it mention "file sample rate does not match device sample rate" ?
• What device is selected in equalizer apo? Is it the default audio device?

ASH resets my selected HpCF profile on startup.

To help me replicate this can you confirm which filter is being applied before you restart ASH? Headphone database, brand/type, headphone, sample/dataset.

It would probably make it easier if you can provide a copy of the settings file. Go to additional tools & settings -> inputs -> settings and presets -> open folder -> settings.ini

[ad2003]

Hi! I appreciate your quick response! Thank you!

Here is what I tried:

Audio Device and Sample Rate Details

Default audio device (Windows Sound Control Panel): 44.1 kHz / 48khz

Default audio device sample rate (Windows Sound Control Panel): 44.1 kHz / 48khz

Default audio device (ASH → Quick Config → Audio Device Format): 44.1 kHz / 48khz

Default audio device sample rate (ASH → Quick Config → Audio Device Format): 44.1 kHz / 48khz

Selected sample rate (ASH → Quick Config → IR Format): 44.1 kHz / 48khz

Are filters/parameters reapplied after changing the sample rate?

Not that I’m aware of.

Equalizer APO Status

When opening the Equalizer APO Configuration Editor and checking the details for
ASH-Outputs\E-APO-Configs\ASH_Toolset_Config.txt, it does report:
“file sample rate does not match device sample rate”.

Additional Context

I'm using KZ ZST X earphones through a Ugreen USB sound card. There seem to be two different filter versions available (“compilation” and “ASH filters”), and there appears to be a mismatch between 44.1 kHz and 48 kHz sample rates.

The file ..\HpCFs\FIRs\KZ\KZ_ZST_X_Sample_A.wav (44.1 kHz) sounds best when my audio device is set to 44.1 kHz, even though Equalizer APO still reports a mismatch.

When I switch my audio device to 48 kHz, the file ..\HpCFs\FIRs\in-ear\KZ_ZST_oratory1990.wav sounds best—again with a mismatch warning.

Disabling the filter results in an audible change, so something is definitely being applied, despite the mismatch warnings.

I also tried the same setup with 48khz - the problem I have right now is that I have a bluetooth set (Soundcore A40) which only supports 44.1khz for some reasons - and the cable version KZ ZSTX - which supports 44.1khz, 48khz and 96khz.

Maybe I could setup a profile for both.

"To help me replicate this can you confirm which filter is being applied before you restart ASH? Headphone database, brand/type, headphone, sample/dataset."

I am applying the KZ ZSTX filters in Compilation / ASH Filters.
It's not always happening, that it resets my selection, but sometimes.

Here is my settings:

`[DEFAULT]
path = C:\Program Files\EqualizerAPO\config
sampling_frequency = 44.1 kHz
bit_depth = 24 bit
brir_headphone_type = Over/On-Ear Headphones - High Strength
brir_hrtf = H8_44K_16bit_256tap_FIR_SOFA
spatial_resolution = Low
brir_room_target = ASH Target
brir_direct_gain = 2.0
acoustic_space = Audio Lab A
version = 3.6.2
fir_hpcf_exp = True
fir_st_hpcf_exp = True
eapo_hpcf_exp = True
geq_hpcf_exp = True
geq_31_exp = True
hesuvi_hpcf_exp = True
dir_brir_exp = True
ts_brir_exp = False
hesuvi_brir_exp = False
eapo_brir_exp = False
sofa_brir_exp = False
auto_check_updates = False
force_hrtf_symmetry = Disabled
er_delay_time = 0.0
mute_fl = False
mute_fr = False
mute_c = False
mute_sl = False
mute_sr = False
mute_rl = False
mute_rr = False
gain_oa = 7.499997138977051
gain_fl = 0.0
gain_fr = 0.0
gain_c = 0.0
gain_sl = 0.0
gain_sr = 0.0
gain_rl = 0.0
gain_rr = 0.0
elev_fl = 0
elev_fr = 0
elev_c = 0
elev_sl = 0
elev_sr = 0
elev_rl = 0
elev_rr = 0
azim_fl = -40
azim_fr = 40
azim_c = 0
azim_sl = -90
azim_sr = 90
azim_rl = -135
azim_rr = 135
upmix_method = Method B
side_delay = 0
rear_delay = 0
show_hpcf_history = False
tab_selected = 122
enable_hpcf = False
auto_apply_hpcf = True
hpcf_current =
hpcf_selected = Anker SoundCore Space A40 Sample A 44.1 kHz 24 bit
enable_brir = True
brir_set_current = B&K-4128, Theatre, 2.0dB, rtings-Target-Low-End, LFR-D-90, In-Ear-Low
brir_set_selected = B&K-4128 Theatre 2.0dB rtings-Target-Low-End In-Ear-Low 44.1 kHz 24 bit Disabled 0.0 90 Low-frequency Response D True True Auto Select
channel_config = 7.1 Surround
prevent_clip = Align Mid Frequencies
qc_brir_headphone_type = In-Ear Headphones - Low Strength
qc_brir_room_target = rtings Target - Low End
qc_brir_direct_gain = 2.0
qc_acoustic_space = Theatre
qc_brand = Anker
qc_headphone = Anker SoundCore Space A40
qc_sample = Sample A
brir_hrtf_type = User SOFA Input
brir_hrtf_dataset = N/A
qc_brir_hrtf = B&K Type 4128
qc_brir_hrtf_type = Dummy Head / Head & Torso Simulator
qc_brir_hrtf_dataset = B&K-4128
sofa_exp_conv = GeneralFIR
qc_crossover_f_mode = Custom Value
qc_crossover_f = 90
qc_sub_response = Low-frequency Response D - Studio, fast & smooth
qc_hp_rolloff_comp = True
qc_fb_filtering = True
crossover_f_mode = Auto Select
crossover_f = 120
sub_response = Low-frequency Response A
hp_rolloff_comp = False
fb_filtering = False
fb_filtering_mode = False
qc_fb_filtering_mode = True
hrtf_list_favs = ["No favourites found"]
hrtf_polarity = Auto Select
tab_bar = 140
fde_hpcf_brand = earbud
fde_hpcf_headphone = AfterShokz Aeropex
fde_hpcf_sample = Rtings
fir_hpcf_toggle = True
fir_st_hpcf_toggle = True
hesuvi_hpcf_toggle = True
geq_hpcf_toggle = True
geq_31_hpcf_toggle = True
fde_hpcf_active_database = Compilation
fde_brir_hp_type = Over/On-Ear Headphones - High Strength
fde_room_target = ASH Target
fde_direct_gain = 2.0
fde_direct_gain_slider = 2.0
fde_acoustic_space = Audio Lab A
fde_brir_hrtf_type = User SOFA Input
fde_brir_hrtf_dataset = N/A
fde_brir_hrtf = H8_44K_16bit_256tap_FIR_SOFA
fde_brir_spat_res = Low
fde_crossover_f_mode = Auto Select
fde_crossover_f = 120
fde_sub_response = Low-frequency Response A - Audio Lab, smooth
fde_hp_rolloff_comp = False
fde_fb_filtering = False
dir_brir_toggle = True
ts_brir_toggle = False
sofa_brir_toggle = False
hesuvi_brir_toggle = False
multi_chan_brir_toggle = False
fde_wav_sample_rate = 44.1 kHz
fde_wav_bit_depth = 24 bit
qc_toggle_hpcf_history = False
qc_hpcf_brand = KZ ZST
qc_hpcf_headphone = KZ ZST
qc_hpcf_sample = oratory1990
qc_e_apo_curr_hpcf = KZ ZST, oratory1990
qc_e_apo_sel_hpcf = KZ ZST, oratory1990, 44.1 kHz, 24 bit
qc_auto_apply_hpcf_sel = True
e_apo_hpcf_conv = True
qc_hpcf_active_database = Compilation
qc_lf_analysis_toggle = True
qc_wav_sample_rate = 44.1 kHz
qc_wav_bit_depth = 24 bit
qc_e_apo_curr_brir_set = B&K-4128, Theatre, 2.0dB, HATS-Target, LFR-D-90, In-Ear-Low
qc_e_apo_sel_brir_set = B&K-4128 Theatre 2.0dB HATS-Target In-Ear-Low 44.1 kHz 24 bit Disabled 0.0 90 Low-frequency Response D - Studio, fast & smooth True True Auto Select no_averaged_meta
e_apo_brir_conv = True
e_apo_audio_channels = 7.1 Surround
e_apo_upmix_method = Method B
e_apo_side_delay = 0
e_apo_rear_delay = 0
e_apo_az_angle_fl = -40
e_apo_az_angle_fr = 40
e_apo_az_angle_c = 0
e_apo_az_angle_sl = -90
e_apo_az_angle_sr = 90
e_apo_az_angle_rl = -135
e_apo_az_angle_rr = 135
e_apo_prevent_clip = Disabled
e_apo_mute_fl = False
e_apo_mute_fr = False
e_apo_mute_c = False
e_apo_mute_sl = False
e_apo_mute_sr = False
e_apo_mute_rl = False
e_apo_mute_rr = False
e_apo_gain_oa = 0.0
e_apo_gain_fl = 0.0
e_apo_gain_fr = 0.0
e_apo_gain_c = 0.0
e_apo_gain_sl = 0.0
e_apo_gain_sr = 0.0
e_apo_gain_rl = 0.0
e_apo_gain_rr = 0.0
e_apo_elev_angle_fl = 0
e_apo_elev_angle_fr = 0
e_apo_elev_angle_c = 0
e_apo_elev_angle_sl = 0
e_apo_elev_angle_sr = 0
e_apo_elev_angle_rl = 0
e_apo_elev_angle_rr = 0
hesuvi_elev_angle_fl = 0
hesuvi_elev_angle_fr = 0
hesuvi_elev_angle_c = 0
hesuvi_elev_angle_sl = 0
hesuvi_elev_angle_sr = 0
hesuvi_elev_angle_rl = 0
hesuvi_elev_angle_rr = 0
hesuvi_az_angle_fl = -30
hesuvi_az_angle_fr = 30
hesuvi_az_angle_c = 0
hesuvi_az_angle_sl = -90
hesuvi_az_angle_sr = 90
hesuvi_az_angle_rl = -135
hesuvi_az_angle_rr = 135
check_updates_start_toggle = True
hrtf_polarity_rev = Auto Select
qc_brir_hp_type = In-Ear Headphones - Low Strength
qc_room_target = HATS Target
qc_direct_gain = 2.0
qc_direct_gain_slider = 2.0

`

Thank you a lot!

[ShanonPearce]

Thanks for the context. I had another look.

• I found one issue relating to resampling accuracy. When selecting a higher sampling rate (base rate for filters is 44.1Khz), the filter will be resampled and the resampling function results in slight differences/ loss of accuracy. It is only a slight difference but could be audible. I will improve this in the next release. Just to confirm did you find audible differences when comparing the different sample rates?
• When you see “file sample rate does not match device sample rate” in equalizer APO that indicates it cant perform the convolution with the filter due to file vs device sample rate mismatch. You would see a flat line in the analysis panel similar to below meaning no filtering is performed. Can you confirm that is the case?

• Can you also confirm is the device selected in equalizer APO the default audio device? You can check in the equalizer apo configuration editor, up the top where you select device (it will say Default (...)). The audio device format in ASH only reports the default audio device so if equalizer APO is running on a different audio device then that (mismatch) flag might not be relevant in that case. As long as the sample rate in ASH matches the sample rate of the device selected in equalizer APO there shouldnt be any problems.

• When you change the sample rate in ASH, apply selection & apply parameters need to be clicked again so that it writes the new wav files. If only one is applied again you could end up with misaligned sample rates (e.g. headphone correction at 44.1khz and binaural at 48khz) which would partially fail. Just confirming that isnt happening?

• Since only one of your devices supports higher sample rates you can leave ASH at 44.1kHz since there should be no benefit of using higher sample rates in terms of quality.

• "ASH resets my selected HpCF profile on startup" The only way I can replicate this issue is when changing the headphone database (e.g. from ASH filters to compilation) before closing the app. The next time it launches it cant find the previously applied filter in that database so it reverts to first in the list. I will fix this one too. Could that be what happened to you?

[ShanonPearce]

I found one more bug where in certain cases the selected sample rate wont be applied when clicking 'apply selection' under headphone correction. It happens in cases where the filters were already applied at one sample rate then a different headphone is selected and sample rate is changed before returning to the previous headphone. I will fix this one too.

[ad2003]

Hi!

1. I didn’t really hear any clear or obvious audible differences when switching between sample rates. At most, I had a slight impression that the sound felt a bit dull at higher sample rates, but I’m not confident whether that was an actual change or just subjective perception.

2. Yes, when the “file sample rate does not match device sample rate” warning appeared in Equalizer APO, the analysis panel showed a completely flat line, indicating that no filtering was applied when using: ASH Filters KZ ZST X -> Sample A; while using ASH Filters KZ ZSTX (withou space) -> Sample A it has no flatline but sound "dull". The highs are missing. Sounds better diabled.

3. Yes, in the Equalizer APO Configuration Editor, the device is set to the “Default” audio device.

4. When changing the sample rate in ASH, I made sure to click both “Apply Selection” and “Apply Parameters” each time, so the filters should have been correctly applied.

5. Yes, the HpCF profile reset on startup happened because I changed the headphone database before closing the app. It seems to happen when switching between "Compilation" and "ASH Filters".

6. I haven’t been able to confirm this one reliably. Switching between my headphones is a bit tricky because one pair is cabled and the other is Bluetooth, so I’m not sure if I encountered this specific bug.

Hope this helps!

Thank you!!