How-to install moduls under Windows.

[erixx11]

Hi Michael,
sitting here and after installing the latest v1.9.76, wondering how to add stuff, now that it's not included.
I am not a programmer, just a translator.
The readme doesn't go into details for us. Maybe the easiest way is a command line for the full install? (I have a selfbuild Windows computer with no space limitations.)

(Personally I want to try Whisper dictation to see if it is at least as good as Dragon. It needs auto-capitalization of the first letter in each sentence, doing it manually each sentence is not handy at all. )

Of course BIG THANKS for all you do, and so fast!

[michaelbeijer]

Hi there!

How did you install v1.9.76?

• Pull from GitHub?
• Download repo as zip?
• pip install supervertaler (in a terminal)?
• downloaded and installed a Windows .exe Release?

If it's just dictation of running text you are after, so no voice commands, I think some of the best options are currently:

• https://wisprflow.ai/
• https://superwhisper.com/

Whisper dictation in Supervertaler is still very rough. It work pretty well though.

Michael

[erixx11]

Hi! Thank you, Michael! Ok, no problem using external tools. Currently I am using Whispertyping, which is nice, except for reviewing and changing segments, because it always capitalizes first word of everything anywhere. I have started translating with Supervertaler and my imported TMX*, and I am enjoying it. Importing a 150.000 unit TMX was slow, with freezing. Then, also batch translating the source text (not long) with it was slow. Just reporting, not complaining :) Good nite Erik

…

On Saturday, January 3rd, 2026 at 22:51, Michael Beijer ***@***.***> wrote: Hi there! How did you install v1.9.76? - Pull from GitHub? - Download repo as zip? - pip install supervertaler (in a terminal)? - downloaded and installed a Windows .exe Release? If it's just dictation of running text you are after, so no voice commands, I think some of the best options are currently: - https://wisprflow.ai/ - https://superwhisper.com/ Whisper dictation in Supervertaler is still very rough. It work pretty well though. Michael — Reply to this email directly, [view it on GitHub](#91 (comment)), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/B2DP3YMN4RRJQBMTYM3SE5T4FA2UJAVCNFSM6AAAAACQTOPEBSVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKMZZHE2DONA). You are receiving this because you authored the thread.Message ID: ***@***.***>

[michaelbeijer]

Hi Erik,

It's great to hear people are starting to use Supervertaler! It is still very rough around the edges and developing quickly, so there are bound to be tons of bugs. If you run into any yourself feel free to let me know, either here in the discussions forum, or at https://github.com/michaelbeijer/Supervertaler/issues. Or just email me at info@michaelbeijer.co.uk!

I made a quick video, demonstrating how dictation works in the grid:

https://www.youtube.com/watch?v=TdbnxhIRLtQ

While making it I had forgotten that you actually don't want the first letter capitalised. I'm not sure, but I think this might be something I could tweak somehow. Note that in the demo video I am using OpenAI Whisper, via the API. It is also possible to use Whisper locally, but this is much slower, unless you have an amazingly powerful computer.

Incidentally, after using Dragon for many years, I now use Talon Voice myself, which is fantastic at voice commands, but not so great at dictating running text. I actually own the latest version of Dragon, but the program is just too darned heavy on my computer. It will often work great for a while, and then start doing really annoying things.

Best wishes,
Michael

[erixx11]

Hi again, a quick note before going to bed. The capitalization in the Whisper family of applets is like "one rule for all". My workflow includes both dictating full new segments: capital 1st letter wanted! But also editing pretranslated segments: changing words in the middle of a segment: here a cap is unwanted! I have been using Dragon with Deja Vu X3 for a decade and they are clever enough to do this automatically. Months ago I contacted the (Dutch) author of WhisperTyping and he promised to look into this, but its caused by the AI itself, it seems and I don't see a fix. Bona nit Erik

…

On Sunday, January 4th, 2026 at 00:48, Michael Beijer ***@***.***> wrote: Hi Erik, It's great to hear people are starting to use Supervertaler! It is still very rough around the edges and developing quickly, so there are bound to be tons of bugs. If you run into any yourself feel free to let me know, either here in the discussions forum, or at https://github.com/michaelbeijer/Supervertaler/issues. Or just email me at ***@***.***! I made a quick video, demonstrating how dictation works in the grid: https://www.youtube.com/watch?v=TdbnxhIRLtQ While making it I had forgotten that you actually don't want the first letter capitalised. I'm not sure, but I think this might be something I could tweak somehow. Note that in the demo video I am using OpenAI Whisper, via the API. It is also possible to use Whisper locally, but this is much slower, unless you have an amazingly powerful computer. [image.png (view on web)](https://github.com/user-attachments/assets/ac213539-4ec5-4ef1-8164-c121b0611678) Incidentally, after using Dragon for many years, I now use [Talon Voice](https://talonvoice.com/) myself, which is fantastic at voice commands, but not so great at dictating running text. I actually own the latest version of Dragon, but the program is just too darned heavy on my computer. It will often work great for a while, and then start doing really annoying things. Best wishes, Michael — Reply to this email directly, [view it on GitHub](#91 (comment)), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/B2DP3YLNB22APSVMFVPYSG34FBINLAVCNFSM6AAAAACQTOPEBSVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKNBQGA2TAMA). You are receiving this because you authored the thread.Message ID: ***@***.***>