Subtitle Edit v5.0.0 is here!

[niksedk]

Release v5.0.0

Subtitle Edit 5 is a major new release and a big step for the project. For the first time, Subtitle Edit runs natively on Windows, macOS, and Linux from a single, modern, cross-platform codebase.

On macOS and Linux the needed media components (mpv/ffmpeg) are bundled in.

---

⚠️ Please read before upgrading

Subtitle Edit 5 is a new application, not just an update of Subtitle Edit 4. It has been rebuilt from the ground up to be cross-platform, so:

• It is not 100% the same app: The look, layout, and some workflows have changed. Some things are in different places, and a few behave differently than in SE4.
• Not every SE4 feature exists in SE5 yet: SE5 covers all the core editing, conversion, sync, video playback, OCR, and online services, but some of the more specialized SE4 tools are not available yet. Features will continue to be added.

💡 Pro-Tip: If you rely on a specific SE4 feature that is missing, please keep SE4 installed alongside SE5. The easiest way to run both side by side is to use the Portable versions of SE4 and SE5, which keep their settings separate and do not interfere with each other.

---

🤔 Which version should I use?

Version	Recommended For	Notes
Subtitle Edit 5	Windows 10 (22H2) or newer, macOS 12+, and Linux.	Recommended for most users.
Subtitle Edit 4	Older Windows versions (Windows 7/8), or older/slower computers.	Continue to use SE4 if SE5 does not run well. SE4 remains available and supported for these cases.

To run SE4 and SE5 at the same time, use the Portable versions—you can try SE5 while keeping SE4 as a fallback.

---

💬 Feedback & Support

Feedback is welcome on the issue tracker—it directly helps prioritize what comes next.
https://github.com/SubtitleEdit/subtitleedit/issues
Please keep reporting issues with nice, detailed descriptions and steps/files to reproduce :)

Thank you to everyone who has supported and contributed to Subtitle Edit over the years! 😊
Especially all the incredible translators who keep SE accessible globally, @mjuhasz for a massive number of PRs (including a ton of Mac fixes for the SE5), and everyone testing the pre-releases, reporting bugs, and helping fellow users in the community. This project wouldn't be where it is today without you!

[niksedk]

Again, it is fine to stay on SE4 as mentioned above :)

[Ouwerkerk78]

would be nice if i can use my TMDB API to translate stuff,

[nawarovec-hash]

1

I really like the new interface, and I wish I can use it, but I can't with these issues.
SE5 doesn't allow live spell-check in the text box. It underlines the wrong word in the grid, and we can't see the word if the line is too long thus, we need to do a spell check at the end to make sure.
Regarding RTL, I've noticed something strange in the new version. When I use "ctrl + ➡" in the RTL language text box, the pointer goes in the opposite direction to the next word instead of the previous one. Also, in translation mode, the original text box shows under the translation column and on the right-hand side while in SE4 everything was aligned perfectly under its column and in a visually sensible fashion.

[OmrSi]

Hey, Nikolaj.

Congrats on the release of SE5, a new milestone!

I look forward to using it, seeing it improve, and bringing you some bug reports. :D

Thank you for all the effort you've put into this; a total rewrite is never easy.

[Rouzax]

Will start testing with SE5.
Does it still support the same CLI commands like:
SubtitleEdit.exe /convert "C:\Path\To\File.srt" subrip /inputfolder:"C:\Path\To\Source" /overwrite /MergeSameTexts /RemoveTextForHI /FixCommonErrors /FixCommonErrors

And will it still honor command flags in order and times of use?

[Rouzax]

One followup question, how are the options determined or set?
Previously you could open the GUI and set all the required settings for FixCommonErrors and RemoveTextForHI which options you wanted

Will start testing soon, many thanks!

[Rouzax]

@niksedk
Follow up on my earlier question about flags being honored "in order and times of use":

Did some testing with seconv 5.0.0 on Linux and it looks like repeating a flag does not run it twice.

In SE4 batch I always ran FixCommonErrors twice on purpose:
SubtitleEdit.exe /convert ... /MergeSameTexts /RemoveTextForHI /FixCommonErrors /FixCommonErrors
The second pass cleaned up new issues the first pass introduced (line merges/splits, timing).

In seconv, passing it twice gives the exact same output as once:

seconv in.srt subrip --overwrite --fix-common-errors
seconv in.srt subrip --overwrite --fix-common-errors --fix-common-errors

Both are byte identical, and the Operations summary only lists FixCommonErrors once, so it gets deduped to a single pass.

It does matter though. Running seconv twice on the same file still changes a few lines vs a single pass (2 and 6 lines on two real ~5-7k line subs I tested, mostly display time / overlap fixes that re-trigger after the first pass).

Would it be possible to run FCE more than once in a single call? Either honor the repeated flag, or maybe a --fix-common-errors-passes:2 option, or loop until stable. For now I just call seconv twice to match the old behavior.

Many thanks!

[Rouzax]

@niksedk, Thanks for the quick pickups.

[niksedk]

Thx for reporting the issues :)

[Rouzax]

@niksedk
Thanks again for the quick turnaround on the other seconv fixes!

One more question while I'm here: I noticed FixCommonErrors in seconv intentionally excludes FixCommonOcrErrors (the source comment says it needs the UI-side IOcrFixEngine + spell-check setup). For cleaning subtitles that were OCR'd from image formats, that dictionary / OCR-replace-list pass is the part that fixes the leftover 0/l/rn style mistakes, so it would be really useful to have headless.

Is the omission a fundamental blocker, or mainly that the spell-check/dictionary engine just is not wired into the CLI yet? Could seconv take an optional dictionary path (e.g. point it at the Subtitle Edit Dictionaries folder / an OCRFixReplaceList) and run that pass, even without the full UI spell-checker? Happy to open a feature request if that is the better place for it.

Thanks!

[andor1999]

Maybe I'm blind, but where are these controls or the equivalent in 5.0? I've gotten used to using them to quickly jump from one sub to the other while listening.
Is this the appropriate place to ask questions like this? It's not exactly a bug or feature request.

[niksedk]

Yes, it's an appropriate place to ask questions :)

The buttons are now below the waveform:

The are now very customizable, see Settings, Waveform/spectrogram, Edit toolbar...

Here you can choose the buttons you want, the order, and size:

Let me know if you are missing any buttons here.

[andor1999]

There is no play previous button and the play next behavior is different from 4.x.
4.x: Play would play the selected line once (or multiple depending on auto-repeat)
Previous: Jumps to the previous line, then same behavior.
Next: Jumps to the next line, then same behavior.

5.1.0: Play action of 4.x replaced by circular arrow. Repeats line continuously.
I can live with that, although there may be a setting to change the number of loops.
Turning off repeat in settings turned out to be the best option for me.
But if I click on circle to get the line looping and then click on Play Next it jumps to the next line and and continues to play without stopping and looping. Essentially, play next is a play button.
And there is no Play Previous.
If you could duplicate the behavior of those 4.x buttons it would be much appreciated.
I use them quite often in my workflow.

[niksedk]

v5.1.0-beta1 is out: https://github.com/SubtitleEdit/subtitleedit/releases/tag/v5.1.0-beta1
Some minor bug fixes + a lot of SE4 parity stuff: https://raw.githubusercontent.com/SubtitleEdit/subtitleedit/refs/heads/main/change-log.txt

[andor1999]

Excellent. I knew the answer was in the settings somewhere, but there are so many. lol Liking the new version. Very snappy. Having the OCR results alongside the originals is something I never knew I wanted. Great improvement. I'm sure I'll find many more. Thanks for all your hard work.

…

On Fri, Jun 26, 2026 at 9:16 AM Rouzax ***@***.***> wrote: Thanks again for the quick turnaround on the other seconv fixes! One more question while I'm here: I noticed FixCommonErrors in seconv intentionally excludes FixCommonOcrErrors (the source comment says it needs the UI-side IOcrFixEngine + spell-check setup). For cleaning subtitles that were OCR'd from image formats, that dictionary / OCR-replace-list pass is the part that fixes the leftover 0/l/rn style mistakes, so it would be really useful to have headless. Is the omission a fundamental blocker, or mainly that the spell-check/dictionary engine just is not wired into the CLI yet? Could seconv take an optional dictionary path (e.g. point it at the Subtitle Edit Dictionaries folder / an OCRFixReplaceList) and run that pass, even without the full UI spell-checker? Happy to open a feature request if that is the better place for it. Thanks! — Reply to this email directly, view it on GitHub <#11744?email_source=notifications&email_token=BEPNKX2GKDGZXZTWSMAFND35BZZTLA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZUGQ2TKMRSUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSWGM33PORSXEX3DNRUWG2Y#discussioncomment-17445522>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BEPNKX4ERDF4DDGNOLVOJWD5BZZTLAVCNFSNUABHKJSXA33TNF2G64TZHMYTMNBXGM2TQNJ3IRUXGY3VONZWS33OHMYTAMZQGA2TSMVBOYBA> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/BEPNKX6BXG6G57IW5ZETBT35BZZTLA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZUGQ2TKMRSUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSVGM33PORSXEX3JN5ZQ> and Android <https://github.com/notifications/mobile/android/BEPNKX33ILYBEVMVUDHEB2T5BZZTLA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZUGQ2TKMRSUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSXGM33PORSXEX3BNZSHE33JMQ>. Download it today! You are receiving this because you commented.Message ID: ***@***.*** .com>

[Hakunamatata67]

I'm testing v5.1.0-beta1.
When I set Speed at 2.0x, the video plays at 2x speed. But then when I go to Settings change something, go back to the main window, the video plays at 1x speed, despite the setting being at 2x speed. I have to chose another speed, then chose 2x again, then it works.

[yolestat]

Feature Request: Smart subtitle splitting using sentence punctuation with proportional timing

Summary

I would like to suggest adding an optional feature to improve subtitle splitting when working with long subtitles, especially those generated by speech-to-text engines such as Whisper, Faster-Whisper, WhisperX and similar tools.

The current splitting algorithm does a good job of balancing subtitle lengths, but it does not always preserve natural sentence boundaries. This often requires manual correction after transcription.

Proposed option

Add a new checkbox in Tools → Split long lines:

☐ Prefer sentence punctuation when splitting subtitles

When enabled, Subtitle Edit would first try to split subtitles at sentence-ending punctuation before applying the standard balancing algorithm.

Supported punctuation could include:

• .
• ?
• !
• ...
• …

Optionally, secondary punctuation such as ; or : could also be considered if no sentence-ending punctuation is available.

Suggested algorithm

When a subtitle exceeds the configured limits:

1. Search for a suitable sentence-ending punctuation mark.
2. If multiple candidates exist, select the one that produces the most balanced subtitle durations.
3. If no appropriate punctuation is found, fall back to the current splitting algorithm.

This ensures that the current behavior is preserved whenever no natural split point exists.

Timing distribution

The timing should not simply be divided equally.

Instead, the original subtitle duration should be redistributed proportionally according to the amount of spoken content (characters, words, or another suitable metric).

Example:

Original subtitle

00:00:00,000 --> 00:00:06,000

I don't know what happened yesterday. Everything changed after dinner.

Result

00:00:00,000 --> 00:00:03,100

I don't know what happened yesterday.

00:00:03,100 --> 00:00:06,000

Everything changed after dinner.

This preserves both the natural sentence structure and realistic subtitle timing.

Benefits

• Produces more natural subtitles.
• Preserves complete sentences.
• Reduces manual editing after AI transcription.
• Improves readability.
• Works especially well with speech-to-text generated subtitles.
• Completely backward compatible, since the option would be disabled by default.

I believe this small optional feature would significantly improve Subtitle Edit's workflow for users working with modern AI transcription engines while leaving the current behavior unchanged for everyone else.

Thank you for considering this suggestion.

[philou-felin]

I never REALLY used previous version, but I started using SE5 yesterday because had PGS files to convert to text. Then I discovered speech-to-text, for commentary tracks than lack subtitles. Those tools are awesome! I can't wait to see how far both nOCR and Whisper CPP (and new models!) go in the future! I think that many improvements are to come for speech-to-text, especially for comments.

[andor1999]

Lost all my SE5 custom setting after a computer freeze and forced shut-off.
It reset everything to defaults.
I don't think it was SE5 that caused the crash, this happens to me infrequently.
It just happened to be at the start of a Fast Whisper speech to text session.
I just thought I'd bring it up because this would suck after someone did extensive customizations.
Is there a way to save custom settings to a file so they can be restored?
Or, better yet, multiple custom setting.

[andor1999]

BTW, the new GUI for speech to text and all the new models blew me away!
Fast Whisper and LargeV3 turbo produces the best timing I've ever seen.
Crisp also has excellent alignment and seems to be best for movies with poor audio.
Unrelated to SE5: I'm also having some success with pre-processing the audio with UVR (Ultimate Vocal Remover).
Contrary to its name it can also remove everything except the vocals.
How much audio pre-processing do the models do?

[andor1999]

I forgot to mention I'm just running an i7-10700 CPU w 16GB.
No GPU. (I wish!)

[philou-felin]

To be honest, I blindly used the "Large-V3" model on my CPU yesterday, but discovered the joy of "large-V3-turbo-q5_0" running on my GTX 970 using the "cuBLAS" backend. This is insanely fast and accurate, so far... Wow! (I use it for commentary tracks: Pan's Labyrinth and The Haunting of Bly Manor.)

[Ouwerkerk78]

Would be really great and nice if TMDB could be added as API translation, from https://www.themoviedb.org the translation of google even with Personal APi key is just as worse as navigating with google maps

[Ouwerkerk78]

So where can i get the api for that LArge-V3 model? Op zo 28 jun 2026 om 18:24 schreef philou-felin ***@***.***>:

…

To be honest, I blindly used the "Large-V3" model on my CPU yesterday, but discovered the joy of "large-V3-turbo-q5_0" running on my GTX 970 using the "cuBLAS" backend. This is insanely fast and accurate, so far... Wow! (I use it for commentary tracks: *Pan's Labyrinth* and *The Haunting of Bly Manor.)* — Reply to this email directly, view it on GitHub <#11744?email_source=notifications&email_token=BRDO6QRS4QOMNP4RAVFZ2I35CFBDVA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZUGYZDMMZZUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSWGM33PORSXEX3DNRUWG2Y#discussioncomment-17462639>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BRDO6QTHBHTMODQIMW3OWXD5CFBDVAVCNFSNUABHKJSXA33TNF2G64TZHMYTMNBXGM2TQNJ3IRUXGY3VONZWS33OHMYTAMZQGA2TSMVBOYBA> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/BRDO6QSM6OXA7INLN7VDMNL5CFBDVA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZUGYZDMMZZUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSVGM33PORSXEX3JN5ZQ> and Android <https://github.com/notifications/mobile/android/BRDO6QUIFXYJHILZLJNXODT5CFBDVA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZUGYZDMMZZUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSXGM33PORSXEX3BNZSHE33JMQ>. Download it today! You are receiving this because you commented.Message ID: ***@***.*** .com>

[philou-felin]

Sorry, I'm just a regular end user. I don't really know how to use an API, let alone package it or whatever. 😅

[GrampaWildWilly]

I'm going to make a guess here. I'm not sure if this is what you're looking for but . . .

If this isn't what you're looking for, oh well. I tried.

[Ouwerkerk78]

Hmm could be yes, i dont have that many api servers, maybe a reinstall could help me get on my way. Thanks 👍🏻 Op ma 29 jun 2026, 02:34 schreef GrampaWildWilly ***@***.***>:

…

I'm going to make a guess here. I'm not sure if this is what you're looking for but . . . 01.png (view on web) <https://github.com/user-attachments/assets/0663a788-c656-4b92-bef1-f665171db0df> If this isn't what you're looking for, oh well. I tried. — Reply to this email directly, view it on GitHub <#11744?email_source=notifications&email_token=BRDO6QSFDZT7N55AYJ74KCT5CG2RVA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZUGY2TAMBXUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSWGM33PORSXEX3DNRUWG2Y#discussioncomment-17465007>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BRDO6QQOXQOKD4FBWBT4HE35CG2RVAVCNFSNUABHKJSXA33TNF2G64TZHMYTMNBXGM2TQNJ3IRUXGY3VONZWS33OHMYTAMZQGA2TSMVBOYBA> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/BRDO6QTF44Z4GVKAMQISFHT5CG2RVA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZUGY2TAMBXUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSVGM33PORSXEX3JN5ZQ> and Android <https://github.com/notifications/mobile/android/BRDO6QQNN5YIRBHKFHDCA435CG2RVA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZUGY2TAMBXUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSXGM33PORSXEX3BNZSHE33JMQ>. Download it today! You are receiving this because you commented.Message ID: ***@***.*** .com>

[dl-73]

I cannot find the dropbox with encoding selection in 5.0.0 version as previously located in the button menu strip in 4.x version. Is it missing, or is it only hidden somewhere? How to save now with different encoding? Thank you for answering.

[niksedk]

I cannot find the dropbox with encoding selection in 5.0.0 version as previously located in the button menu strip in 4.x version. Is it missing, or is it only hidden somewhere? How to save now with different encoding? Thank you for answering.

You can enable encoding in the toolbar here:

[dl-73]

Great thank you. I missed this setting probably.
And thanks for the new version. Really nice and fresh look.

May I also have a suggestion for a new feature? Maybe it is nonsense, but anyway...
I know there is already the Beautify function, but - sometimes when I approach a new downloaded subtitles and they are not accurate, while fixing, it is easier for me to start with a uniform look. I mean it would be nice if the subtitles are kind of rounded. It is then more straightforward when using the small up/down arrows to shift the subtitle a little.
Example:
original: 00:00:23,774 --> 00:00:27,528
after rounding by factor of 20 ms: 00:00:23,780 --> 00:00:27,540
The feature would allow to round all or selection of subtitles by given factor up or down (can be e.g. matching frame rate).

Please scratch the idea if it feels nonsense.

[Jolar]

I just want to say thank you! I got prompted with an update, and saw it was a version jump, but I wasn't expecting anything of this scale. I have only been working five minutes with the new version now, but to my surprise everything felt good and intuitive right away. Nice work!

[niksedk]

A new tool - AI review is included in latest beta: https://subtitleedit.github.io/subtitleedit/features/ai-review.html

[Ouwerkerk78]

It takes me up to 30 minutes to translate SRT with my Nvidia API with new 5.0.1 Op vr 3 jul 2026, 21:47 schreef Dutch-Wildfire ***@***.***>:

…

Thanks for all your efforts, but I'm going to stick with version 4. For me, the new GUI is unworkable. The old GUI is tight, to the point. The new GUI feels way to loose, not coherent. I don't really know how to describe it best. Never the less, that's a me issue. As I said, thanks for all your great work, but for me - I'll stick to version 4 as long as possible. — Reply to this email directly, view it on GitHub <#11744?email_source=notifications&email_token=BRDO6QW2JOO22I7BQGDDGC35DAETTA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZVGI3DOMRRUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSWGM33PORSXEX3DNRUWG2Y#discussioncomment-17526721>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BRDO6QWXTB2U54VHJ4UCBI35DAETTAVCNFSNUABHKJSXA33TNF2G64TZHMYTMNBXGM2TQNJ3IRUXGY3VONZWS33OHMYTAMZQGA2TSMVBOYBA> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/BRDO6QXAGTY6NOR6AT5HFKD5DAETTA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZVGI3DOMRRUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSVGM33PORSXEX3JN5ZQ> and Android <https://github.com/notifications/mobile/android/BRDO6QQIRTTI3WQSQUIRYXD5DAETTA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCNZVGI3DOMRRUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSXGM33PORSXEX3BNZSHE33JMQ>. Download it today! You are receiving this because you commented.Message ID: ***@***.*** .com>

[torvchen]

I've encountered an issue with the "Toggle play/pause" shortcuts (there are two identical?), at least in the latest version. When I pause the video and use the waveform to adjust the subtitle start time by holding Alt and dragging, the "Toggle play/pause" shortcuts stop working until I manually click back into the text box.

[UUMARTIN]

5.x—the more it’s updated, the more problems there are, and the more it seems to be taking a step backward. I don’t know what’s going on.

[niksedk]

And we have Subtitle Edit v5.1.0-rc6 now - do give it a test run :)

[UUMARTIN]

I think SE v5.1.0 closes many of the remaining gaps compared to SE 4.x. The change log since v5.0.0 is huge: https://github.com/SubtitleEdit/subtitleedit/blob/main/change-log.txt

Which features are not working well for you? Please reply with as much detail as possible (steps to reproduce, screenshots, sample subtitle files, error messages, etc.) so we can investigate and fix the issues.

If you prefer the old interface or don't want to learn the new UI right now, that's completely fine—you can continue using SE 4.0.16.

First, thank you for your response. The main issue is....

1. Editing to insert a pre-prepared subtitle file “midway through”—in version 4.x, this was always very simple, as the system would automatically insert and arrange the subtitles according to the timing (for example, when working on a concert video and inserting completed lyrics partway through). I recall that v5.1.0-beta1 seemed to manage this with some difficulty, but now it appears that pre-prepared subtitle files can only be inserted at the very end (in 4.x, inserting them at any point midway through was very convenient).
2. Additionally, in... Selected Line > Speech-to-Text > Input Language... the file language selection is automatically locked. Every time you adjust the recognition settings, you have to manually adjust the target language as well; there’s no longer a feature to remember the last selection (in 4.x, it would remember which target language was used last time).

[niksedk]

Thanks for the detailed feedback - both issues are now fixed and will be in the next RC update.

1. Inserting a pre-prepared subtitle file midway (#12569): "Insert subtitle after current line..." in the subtitle grid context menu now works on any line, not only the last one - and like SE 4.x, the inserted file keeps its own time codes so pre-timed lyrics land at their real positions (it is only shifted if it would overlap the selected line). The waveform context menu also has "Insert subtitle file at video position..." available anywhere on the timeline now - it inserts the file at the right-clicked position and warns before creating overlapping lines.

2. Speech-to-text language not remembered (#12571): the input language now works like SE 4.x again - the last used language is remembered and restored when the window opens, changing the engine or other recognition settings no longer resets it, and the "Selected lines > Speech-to-text" flow no longer overrides your choice with the auto-detected language on every run.

Please give the next build a try and let us know if anything still does not match your workflow :)

[niksedk]

v5.1.0-rc7 is out: https://github.com/SubtitleEdit/subtitleedit/releases
Better?
What language are you using?

[UUMARTIN]

v5.1.0-rc7 is out: https://github.com/SubtitleEdit/subtitleedit/releases Better? What language are you using?

That's great! It looks like everything's all set. Thank you so much for all your hard work.

[andor1999]

I see you've made progress on customizing the Tool Bar under waveform, but you still haven't duplicated the functionality of the buttons under the Translate tab of SE 4.x. The is no previous button, the Next button plays the next sub but keeps on playing instead of stopping.
I don't generally create subs from scratch so the default buttons are not useful. Those old Translate tab functions are my bread and butter.
I can't move to SE 5 completely until those are duplicated.

[andor1999]

Ideally, I'd like those exact buttons. :)

[niksedk]

OK, will be available in next RC, but you will need to configure the toolbar for them to appear

[andor1999]

👍That part I can do. :)
Thanks.

[andor1999]

Installed rc14. The buttons work perfectly now.
Thanks.

[torvchen]

SE5 is becoming better and better. :) Thank you for your hard work!

I wrote earlier about the issue with the "Toggle play/pause" shortcuts not responding on the waveform. Now I've noticed that there are more shortcuts that randomly stop responding. I'm not quite sure in which situations this happens, but I've identified these triggers so far:

• When I use the "Toggle shot changes at video position" shortcut (ctrl+alt+A in my case) on an area where there is no subtitle (with focus either on the waveform or in the text box)
• When I adjust the subtitle start or end time by holding Alt and dragging, many shortcuts stop responding until I click on the text box or the waveform.

At least these shortcuts stop responding (there might be more):

• Toggle play/pause
• Video, one frame back
• Video, one frame forward
• Go to previous shot change
• Go to next shot change

Also, the "Toggle shot changes at video position" doesn't always respond without first clicking on the waveform, i.e. when focus is in the text box.

Ideally, these shortcuts should work regardless of where the current focus is.

Another issue I haven't been able to solve: the shortcut "Extend selected lines to next shot change (or subtitle)" no longer leaves the configured two-frame gap before the shot change. Even though "Beautify Time Codes" is set up to leave a two-frame gap, the shortcut extends all the way to the edge. (In SE4, it correctly left the two-frame gap before the shot change).

I use RC10.

Hope you can have a look at these issues in an upcoming version. I really appreciate your hard work!

[niksedk]

Thanks for the detailed report - the state-by-state observations made these much easier to track down :)

• "Extend selected lines to next/previous shot change" now leaves the configured in/out cues gap again (Leave the configured gap when extending lines to a shot change #12672).
• Two shortcut killers fixed (Stop menus and flyouts from leaving shortcuts dead #12673): a closed menu/context menu could silently keep keyboard focus and swallow every shortcut, and "Toggle shot changes at video position" only fired while the waveform had focus - it now works like the other shot change shortcuts, regardless of focus.

Both will be in the next RC. The Alt+drag case may have an additional cause we have not pinned down yet - if shortcuts still stop responding after Alt+dragging in the next RC, please report back :)

[torvchen]

Wow, that was fast. Much appreaciated. :) I will report back if I notice anything in the next RC.

[torvchen]

Now that I've tested it a bit, I see that most of these problems are solved. Thanks! But as you suspected, the Alt+drag issue is still there. However, it's a minor issue.

Also, I've noticed that alignment shortcuts don't work in the text box. They're listed as subtitle list view shortcuts, so I guess this is intentional? But if possible, it would be great if they worked in the text box, too.

[mogan616]

The updated waveform is great, especially the ability to make it way larger. I like some of the updates but some of the things that are missing are bugging me:
The window no longer shows progress when loading a subtitle from an mkv file, my current workaround is clicking the locked up window so Windows marks it as not responding and that status disappearing is my cue that it's done.
In the "remove text for hearing impaired" window, not having a dropdown with common symbols for the "between" and "contains" options adds work where a click could solve it before.
"Point sync via another subtitle" is a feature I use a lot and the right list not following when you select a sync point in the left nor the ability to use the END or HOME keys on the keyboard to navigate lines makes it a bit cumbersome to use now.

[niksedk]

Thanks for the detailed feedback - all three of these are now addressed or in progress.

Point sync via another subtitle - right list not following

Fixed in 0ae5380 (#12529): clicking a line in the left list now scrolls the other subtitle's list to the matching time, offset by the nearest existing sync point so it still lands right when the two files drift apart. The right list's selection is deliberately left alone, since that's your pick for the next sync point. This landed after 5.0.0, so it's in the 5.1 betas/RCs.

Home/End in point sync

You were right, this was missing - the grids there had no Home/End handling at all. PR #12677 adds it to both point sync windows (and factors the snippet that was copy-pasted across ~30 other windows into a shared helper). Bonus: Home/End in the left list also triggers the scroll-sync above.

MKV progress

This one was more subtle than it looked. Progress was added before 5.0.0 (#6772), but only reached some paths:

• multi-track files opened the track picker first, which parsed on the UI thread with no progress - fixed in MKV parsing progress bar doesn't appear with multiple subtitles #12193, in 5.1 betas
• PGS / Blu-ray tracks never had it at all - that branch called into the Blu-ray reader synchronously on the UI thread with no progress callback, so a large mkv with a PGS track just froze. Since PGS is the usual case for rips, this is most likely what you hit. PR Show MKV progress for PGS tracks and surface extraction errors #12679 fixes it.

If your file was a PGS track, that last one is the fix - worth re-testing on the next beta once it's merged.

The hearing-impaired dropdown suggestion is noted separately.

[jmaraujouy]

Oh, such a pity i didn't read this before that darn winget updated the app! :( I will try to downgrade and pray that my files are still there (rules, interjection lists, etc)... EDIT: Everything is still there... :)

Apart from that, the new version is looking great. But unfortunately there are some features that I need that now are not available in v5, so I'm sticking with 4.x.

[niksedk]

@jmaraujouy v5.1.0 will be out soon... what features are you missing from SE 4?

(Also, you can press Ctrl+4 for SE4 import/look)

[niksedk]

SE v5.1.0 is very close - give the latest RC a spin from here: https://github.com/SubtitleEdit/subtitleedit/releases

[andor1999]

rc14 was blocked by Windows Smart App. I had to turn it off. This is some new protection Microsoft added to Windows.
I don't believe it blocked installation, just the launch.
I tried to dig down to see what part was blocked, but had no luck. The instructions led nowhere.
https://support.microsoft.com/en-US/Windows/Security/Threat-Malware-Protection/smart-app-control-has-blocked-part-of-this-app

[niksedk]

Thanks for the report! This is a reputation-based false positive — Smart App Control blocks new binaries it has no reputation data for, which hits fresh RC builds especially hard.

You can report the false positive to Microsoft here (choose "incorrectly detected"): https://www.microsoft.com/en-us/wdsi/filesubmission. Once Microsoft processes it and the binaries build reputation, the block should go away.

[GrampaWildWilly]

I'm running Windows 11 Pro 64-bit version 25H2 build 26200.8894 with the default built-in Windows Defender. I've never had it squawk about any version of SE, SE 4, SE 5, betas, Previews, Release Candidates, general availability releases. I have not had to tell Defender that SE is safe. I have not had to tell the firewall that SE is safe. It's always just worked. The SE files I've downloaded have never triggered the antivirus & running SE has never triggered the antivirus. But even if it had, I would have simply told Defender don't quarantine this & gotten on with things. False positives are a fact of life with antivirus programs. They are programs, like any other programs: SE, Firefox, Notepad++, Acrobat Reader, Windows Defender . . . They're all just programs. Sometimes, programs have bugs. Antivirus is not immune to bugs. Sometimes they mistakenly flag software you know is safe. You have been using SE for years (I'm guessing). It has proven to be trustworthy. Why would you suddenly start distrusting SE? Don't bother Microsoft. Just tell your antivirus to make an exception for SE & get on with things.

[Hakunamatata67]

The last RC is nearly perfect for my work. Just one request if possible: when right-click on a line, is it possible to put back "Style/Actors" at the top of the menu instead of "Chose dictionary" ?