*Update! This is an absolute disaster! I have no ida what I'm doing, and this thing runs on text files, not sliders, so I can't actually hear things live. I might try another digital option, but honestly, I think I'd be better off with a real mixer with manual controls. The problem is that they probably cost hundreds of dollars! I'll need to see. But the fact that nothing like this is accessible for the blind is not only dishearting, it's downright upsetting. Do we not have the right to work on such things as do the sighted?*
I have returned to my project of frequency restoration in acoustic recordings.
https://www.reddit.com/r/OldOpera/comments/1oqgz2k/frequency_restoration_in_acoustic_recordings/
Because I know very little about all of this, I asked more questions of Perplexity. (This is not medical information or something so important that it requires meticulous manual research, so I do trust Perplexity here.) It seems that, even though programs which explicitly work with frequency restoration are not accessible to the blind, there is another way to do this via equalisers, harmonics, and other means. This is exactly what I was told to do. I would prefer an analogue setup, since I could hear things as I moved the sliders and knobs, but this might be more precise. This also might be more accurate in the longrun, since there is apparently no real way to restore frequencies faithfully i.e. it would be guesswork and who knows how realistic it would sound. This way, it's working with what's already present and simply enhancing it. I am going to use acoustic and electric recordings of Schipa and Bonci as my controls, so that I can judge how early electrics compare directly to acoustics.
"The good news is that you can build a usable, keyboard‑driven EQ and harmonic‑enhancement workflow without relying on mouse‑dependent GUIs. NVDA will let you navigate most of the necessary controls, and you can use a combination of host‑agnostic EQ and a simple harmonic plugin.
1. Accessible EQ setup (Windows‑friendly)
A solid, free option is Equalizer APO running system‑wide, with a simple host such as Audacity (which you already know from your post‑thread) or TagEditor‑style players.
videoproc
+1
Typical flow:
1.
Install Equalizer APO
•
Download from equalizerapo.com and run the installer.
•
At the device‑selection screen, choose only your playback device (e.g. “Speakers” or your USB DAC) and accept the reboot.
•
After reboot, open the Equalizer APO Configurator via Start Menu or the task‑bar icon.
2.
Set up a basic config tab
•
In the Configurator, remove the default tabs if you want a clean slate.
•
Press Tab through the controls until you reach the “+” button; press Enter.
•
Choose “Control” → “Include” from the menu, then select a new text file (e.g. Tamagno.cfg).
•
Return to the main pane, press Alt+Tab to cycle to the new tab; you will now see a blank filter list.
3.
Add a shelving EQ for bass
•
Position on the “+” button again; press Enter.
•
Navigate to “Basic filters” → “PEQ” or “Low‑shelf”.
•
Use the numeric fields to set:
•
Filter type: Low‑shelf
•
Frequency: around 200–250 Hz to start.
•
Gain: +1 dB to +3 dB (very gentle).
•
Repeat with a High‑shelf around 3–5 kHz if you want subtle vocal presence, again only +1 to +2 dB.
4.
Apply and test
•
Go to File → Save in the Configurator.
•
Make sure the power‑toggle is on for your device.
•
Play your Tamagno file in Audacity or any other player and judge the warmth versus “tinniness”.
NVDA will announce labels such as “Gain”, “Frequency”, “Q”, and “Filter list”, so you can tab through each numeric field and adjust by typing the value directly.
equalizerapo
+1
2. Harmonic enhancement (accessible workflow)
For harmonic “body” without heavy distortion, you want a straightforward exciter/saturator plugin. A free, relatively simple option is HarmonicAdder (Anarchy Sound Software), which is a VST plugin that generates overtones from the dominant frequency content.
youtube
reddit
How to use it with an accessible host:
•
Install VST2 into Audacity’s plugin folder (e.g. C:\Program Files (x86)\Audacity\Plug‑Ins).
•
In Audacity:
•
Open your Tamagno file.
•
Press Alt+P → “Add / Remove Plugins…” to ensure the plugin is listed.
•
Press Alt+L to open the “Effect” list, then navigate to the exciter/saturator and select “Apply”.
•
Use a low drive setting (e.g. 10–25%) and a highpass before saturation if your host supports a simple EQ‑in‑chain, so you avoid muddying the bottom.
Because the plugin window is usually a simple dialog, you can tab through parameters such as “Drive”, “Mix”, and “High‑cut”; NVDA will read each label if the plugin is coded accessibly.
github
+1
3. Keyboard‑driven listening‑test schema (for your comparison)
To keep the process systematic, you can adopt a simple keyboard‑driven test set:
1.
File list
•
Keep one directory for each singer: Tamagno, Affre, and one or two electric‑era references (Schipa/Bonci).
•
Use Windows Explorer with NVDA to open each folder, press Enter on one file at a time.
2.
Quick‑switch profile
•
In Equalizer APO, save one profile for Tamagno (slightly warm, cautious +‑bass), one for Affre (neutral), and one for Schipa (closer to the original, electric‑era target).
•
In the Configurator, switch tabs with Ctrl+Tab and then press File → Save after each change.
3.
Test sequence
•
Start playback (e.g. in Audacity or your preferred player).
•
Use Spacebar to play/pause, Shift+→/← to nudge by 5 seconds.
•
For each file, ask:
•
Is the voice more “body‑ful” but still recognisably Tamagno‑like?
•
Are consonants sharper without sounding artificial?
•
Has the timbre begun to resemble your 1926/30 electric reference more than the 1903 original?
Because you are already using a clean transfer, the main variables are low‑shelf gain, high‑shelf presence, and degree of harmonic saturation. You can keep a small text log (e.g. Notepad) with entries like:
•
Tamagno+2.5dB_250Hz_Sat20%
•
Tamagno+1.5dB_200Hz_Sat10%
and move back and forth with the keyboard until you find the most musically convincing setting.
4. Practical suggestions for your goal
Given your stated aim (less tinny, more like early electrics, perhaps 1926–30), I would:
•
Never boost the bass so much that the fundamental pitch of the voice goes soft or “boomy”.
•
Use the harmonic plugin only to add a thin layer of upper‑mids and lower‑highs, not to create a modern‑sounding recording.
•
Keep your original 1903 transfer as a reference file and compare it with the processed version on the same tracklist, using the same signal path and volume.
If you like, I can draft a concrete step‑by‑step checklist tailored to your usual workflow (e.g. starting from Audacity, Equalizer APO, and one exciter plugin), with explicit NVDA‑friendly key sequences and values you can try first."
We'll see what happens and how accessible all of this is.
#AcousticRecordings #FrequencyRestoration #music #phonographs #recordings #records #restoration #sound #technology