Skip to content

v26.10

Latest
Latest

Choose a tag to compare

View all tags
@jonata jonata released this 18 Oct 01:23
· 13 commits to master since this release
v26.10
4ec1b03

Subtitld creates, edits, transcribes, translates and dubs subtitles. A package for every system it runs on is in the assets below.

What's new

  • Dubbing. A synthetic voice for each speaker speaks the subtitles, fitted to their durations if you choose. Export the mix, a track per speaker, a clip per subtitle, or the background alone.
  • Add-ons. Transcription, translation and voices come from add-ons installed inside Subtitld: speech recognition engines (whisper.cpp, Vosk and more, AssemblyAI in the cloud), text-to-speech voices (Piper, Kokoro, Coqui XTTS, F5-TTS and more), offline translation, and voice and music separation. Edge TTS and Google TTS voices stay built in.
  • Record mode. Record speech over the video: each phrase becomes a subtitle as you speak, and an add-on engine fills in its text.
  • Audio panel. EQ, compressor and gate for the voice, the music and each speaker, in a mixer, with an input monitor for recording.
  • subtitld.cc. Find, open, rate and publish subtitles from a panel of their own.
  • Plain-text tab. Edit the subtitles as SRT, Markdown or JSON.
  • Vertical timeline, a tab beside the video, and a keyboard shortcuts panel.
  • 17 interface languages, the system's by default.
  • A new icon and logo.
  • Snapping and gluing at subtitle edges, and a drag (or a held nudge key) undone in one step.

Formats

Subtitld now reads and writes every format itself.

  • New: EBU STL, MacCaption (MCC), CEA-708, YouTube's timed text (srv3), Bilibili's BCC, and the JSON of speech-to-text services (Whisper, WhisperX, Deepgram, AWS, Rev, Google, Azure, Speechmatics and more).
  • From the video itself: its closed captions (CEA-608 and CEA-708) and its subtitle tracks.
  • Rewritten to read what tools write and write what players show: SRT, WebVTT, ASS and SSA, TTML and DFXP, SAMI, SCC, SBV, MicroDVD and USF.

Playback

  • Picture and sound in sync, timed from the sound device itself (on Windows, through WASAPI).
  • Subtitld starts, and plays silently, without a sound device.

Projects (.usfx)

  • An icon of their own, and a preview: a card with the frame where the project was left, its subtitle and the timeline there. It shows in Windows Explorer and, with the AUR package, in GNOME Files.
  • Double-clicked in a file manager, a project opens in Subtitld.

Downloads

System File
Windows Subtitld-…-Setup.exe (installer, for you or for everyone), Subtitld-….msix, or Subtitld-…-portable.zip
macOS (Apple Silicon) Subtitld-….dmg. It is not signed: the first time, allow it in System Settings → Privacy & Security.
Linux Subtitld-…-x86_64.AppImage, Subtitld-….flatpak, subtitld_…_amd64.snap; also in the Snap Store, on the AUR and on PyPI (.whl and .tar.gz)
Haiku subtitld-…-x86_64.hpkg and its Python packages (.hpkg), and subtitld-…-haikuports.tar.gz, the recipes for HaikuPorts

Removed

  • The built-in whisper.cpp and AssemblyAI engines, now add-ons.
  • The built-in Google translation: translation comes from add-ons.

Build version: 26.10.18.0124

The AppImage and the snap were rebuilt on 2026-10-18 as 26.10.18.0446: the first AppImage did not open on systems without libfuse2, which current Ubuntu does not install, and the first snap showed no window on Wayland desktops.

The Haiku packages were rebuilt on 2026-10-18 as 26.10.18.1737: the first ones left projects (.usfx) typed as zip files, without their icon, and the Python packages are now built as haikuports builds its own. subtitld-26.10.18.1737-haikuports.tar.gz holds the recipes for haikuports.

Assets 31
Loading