Hushcue 0.2.5

What's new in 0.2.5:

• Translate to English, a switch in Settings: videos in other languages get English subtitles. The language spoken
  is detected (the log says which) and translated while transcribing; English videos are transcribed as usual.
  Translating uses the large-v3 model for every quality (downloaded the first time, 3.1 GB), because the faster turbo
  model of Fast and Balanced can't translate. Re-run this part and Try another model translate too.

Hushcue 0.2.5 for Windows 10 (1809) and 11, x64: subtitles from any video, made on your own PC.

• Hushcue-0.2.5-setup.exe: installer. Installs for the current user (no admin needed) and can add
  "Generate subtitles with Hushcue" to the right-click menu of video and audio files.
• Hushcue-0.2.5-win-x64.zip: portable copy; unzip anywhere and run Hushcue.exe.

Hushcue needs ffmpeg on PATH (winget install Gyan.FFmpeg). On first run it downloads the speech model
(about 1.6 GB) and, with an NVIDIA GPU, NVIDIA's cuBLAS (about 630 MB); without a GPU it runs on the CPU.
Installed copies update themselves from here; check the files against SHA256SUMS.txt.
The installer is not code-signed yet, so Windows SmartScreen may warn: choose More info, then Run anyway.

Hushcue 0.2.4

What's new in 0.2.4:

• A bigger video while you check subtitles: drag the divider between the video and the subtitle list, or hide the list
  with the new button next to the timeline zoom. The app remembers both. Double-click the divider to reset it.

Hushcue 0.2.4 for Windows 10 (1809) and 11, x64: subtitles from any video, made on your own PC.

• Hushcue-0.2.4-setup.exe: installer. Installs for the current user (no admin needed) and can add
  "Generate subtitles with Hushcue" to the right-click menu of video and audio files.
• Hushcue-0.2.4-win-x64.zip: portable copy; unzip anywhere and run Hushcue.exe.

Hushcue needs ffmpeg on PATH (winget install Gyan.FFmpeg). On first run it downloads the speech model
(about 1.6 GB) and, with an NVIDIA GPU, NVIDIA's cuBLAS (about 630 MB); without a GPU it runs on the CPU.
Installed copies update themselves from here; check the files against SHA256SUMS.txt.
The installer is not code-signed yet, so Windows SmartScreen may warn: choose More info, then Run anyway.

Hushcue 0.2.3

What's new in 0.2.3:

• Preview first, a switch under Result on Generate subtitles: the subtitles are made in a temporary folder and
  nothing is saved next to your videos or in the output folder. Check and edit each preview, then Keep it (as an
  .srt, a subtitle track or burned in, chosen there) or Discard it. Keep all / Discard all on the finished
  screen do every preview at once. Closing the app asks first, then discards previews that weren't kept.
• Preview first replaces the "Review before embedding" setting, which wrote the .srt before you had looked at it.

Hushcue 0.2.3 for Windows 10 (1809) and 11, x64: subtitles from any video, made on your own PC.

• Hushcue-0.2.3-setup.exe: installer. Installs for the current user (no admin needed) and can add
  "Generate subtitles with Hushcue" to the right-click menu of video and audio files.
• Hushcue-0.2.3-win-x64.zip: portable copy; unzip anywhere and run Hushcue.exe.

Hushcue needs ffmpeg on PATH (winget install Gyan.FFmpeg). On first run it downloads the speech model
(about 1.6 GB) and, with an NVIDIA GPU, NVIDIA's cuBLAS (about 630 MB); without a GPU it runs on the CPU.
Installed copies update themselves from here; check the files against SHA256SUMS.txt.
The installer is not code-signed yet, so Windows SmartScreen may warn: choose More info, then Run anyway.

Hushcue 0.2.2

What's new in 0.2.2:

• Edit subtitles opens the subtitle track inside a video (.mkv, .mp4) when there is no .srt next to it: the track is
  saved as an .srt next to the video (name.eng.srt) and opens in the editor. With more than one track you pick
  which. Picture subtitles (Blu-ray PGS, DVD VobSub) can't be edited as text; the app says so and lets you pick an
  .srt instead.
• Clear empties the Recent list on the Edit subtitles page (the files themselves are not touched).

Hushcue 0.2.2 for Windows 10 (1809) and 11, x64: subtitles from any video, made on your own PC.

• Hushcue-0.2.2-setup.exe: installer. Installs for the current user (no admin needed) and can add
  "Generate subtitles with Hushcue" to the right-click menu of video and audio files.
• Hushcue-0.2.2-win-x64.zip: portable copy; unzip anywhere and run Hushcue.exe.

Hushcue needs ffmpeg on PATH (winget install Gyan.FFmpeg). On first run it downloads the speech model
(about 1.6 GB) and, with an NVIDIA GPU, NVIDIA's cuBLAS (about 630 MB); without a GPU it runs on the CPU.
Installed copies update themselves from here; check the files against SHA256SUMS.txt.
The installer is not code-signed yet, so Windows SmartScreen may warn: choose More info, then Run anyway.

Hushcue 0.2.1

What's new in 0.2.1:

• Try another model on part of a file while reviewing: select subtitles (or a gap), choose Try another model,
  and pick Fast, Balanced or Thorough. What that model heard shows next to the subtitles there now, with the words
  that differ marked; play it over the video, then use it (Ctrl+Z puts the old ones back) or keep the current
  ones. A model not on the PC yet (Thorough's large-v3, 3.1 GB) downloads first, if you agree, and can be stopped.
• "Re-run this part" keeps a word another model times just before or after the selected subtitles, instead of
  dropping it.
• For developers: Hushcue.Cli rerun --from S --to S [--preset NAME]... FILE runs a stretch through each preset
  and prints the words that differ.
• Fixed: MKV to MP4 stopped at once with an empty error message, for every file.
• An error with no message of its own now says what kind of error it was, instead of showing an empty box.

Hushcue 0.2.1 for Windows 10 (1809) and 11, x64: subtitles from any video, made on your own PC.

• Hushcue-0.2.1-setup.exe: installer. Installs for the current user (no admin needed) and can add
  "Generate subtitles with Hushcue" to the right-click menu of video and audio files.
• Hushcue-0.2.1-win-x64.zip: portable copy; unzip anywhere and run Hushcue.exe.

Hushcue needs ffmpeg on PATH (winget install Gyan.FFmpeg). On first run it downloads the speech model
(about 1.6 GB) and, with an NVIDIA GPU, NVIDIA's cuBLAS (about 630 MB); without a GPU it runs on the CPU.
Installed copies update themselves from here; check the files against SHA256SUMS.txt.
The installer is not code-signed yet, so Windows SmartScreen may warn: choose More info, then Run anyway.

Hushcue 0.2.0

What's new in 0.2.0:

• Subtitles read like professional ones: lines and subtitles break where the sentence does (after a comma, before
  "and"), not after "the" or between a first and last name, and each subtitle stays up long enough to read.
• Subtitles no longer appear before anyone speaks: word timing is trimmed to the speech itself (in the test video,
  subtitles now start within 0.1 s of the speech 95% of the time, up from 2%).
• Full stops and capitals no longer go missing in long stretches of speech, and fewer words are misheard (in the
  test set, 356 of 372 tricky words right, up from 348).
• "Thank you." and "you" no longer appear over noise, tones or silence; "Thanks for watching" is kept when someone
  really says it.
• The window fits any screen: it opens at the right size for your display scaling, can be made much smaller
  (down to 600×500), and every page rearranges itself for a narrow or short window instead of cutting things off.
• On a narrow window the options move under the file list, the review editor puts the subtitles under the video,
  and buttons wrap onto a second line; a window too short for a page scrolls it.
• The subtitle shown over the video in the review grows and shrinks with the picture.
• Fixed: long file names and counts running under other text, the timeline drawing past its edges, the options
  card stopping short of the right edge, the setup screen's download text overlapping, and the setup card sitting
  at the top of the window instead of in the middle.
• For developers: the app's CLI, tests and engine build on a Mac (CPU only), Windows and macOS builds of the
  engine use separate folders so one checkout can be shared, and a speaking-styles test video is made with macOS
  voices (tools/bench/make_speaking_styles.py); Hushcue.Cli timing scores word timing against it, and
  docs/subtitle-quality-research.md has the research behind this release.

Hushcue 0.2.0 for Windows 10 (1809) and 11, x64: subtitles from any video, made on your own PC.

• Hushcue-0.2.0-setup.exe: installer. Installs for the current user (no admin needed) and can add
  "Generate subtitles with Hushcue" to the right-click menu of video and audio files.
• Hushcue-0.2.0-win-x64.zip: portable copy; unzip anywhere and run Hushcue.exe.

Hushcue needs ffmpeg on PATH (winget install Gyan.FFmpeg). On first run it downloads the speech model
(about 1.6 GB) and, with an NVIDIA GPU, NVIDIA's cuBLAS (about 630 MB); without a GPU it runs on the CPU.
Installed copies update themselves from here; check the files against SHA256SUMS.txt.
The installer is not code-signed yet, so Windows SmartScreen may warn: choose More info, then Run anyway.

Hushcue 0.1.1

What's new in 0.1.1:

• Words that sound like other words (calm / come / cum, their / there, principle / principal, ...) are flagged
  "not sure" in the review when Whisper wasn't sure of them, with one-click buttons to swap in the other spellings.
• Fixed: the Vocabulary setting only reached the first 30 seconds of each file. Names and words in it are now
  spelled your way all the way through (in the test set, names right went from 15 to 31 of 38 on Balanced).
• For developers: Hushcue.Cli eval and a spoken test set (eval/lookalike) measure transcription quality per preset.

Hushcue 0.1.1 for Windows 10 (1809) and 11, x64: subtitles from any video, made on your own PC.

• Hushcue-0.1.1-setup.exe: installer. Installs for the current user (no admin needed) and can add
  "Generate subtitles with Hushcue" to the right-click menu of video and audio files.
• Hushcue-0.1.1-win-x64.zip: portable copy; unzip anywhere and run Hushcue.exe.

Hushcue needs ffmpeg on PATH (winget install Gyan.FFmpeg). On first run it downloads the speech model
(about 1.6 GB) and, with an NVIDIA GPU, NVIDIA's cuBLAS (about 630 MB); without a GPU it runs on the CPU.
Installed copies update themselves from here; check the files against SHA256SUMS.txt.
The installer is not code-signed yet, so Windows SmartScreen may warn: choose More info, then Run anyway.