0.7.0 was the last version on this listing; 0.8.0 to 0.14.0 came out on GitHub only (https://github.com/Multysquid/shisu-ko/releases has every release's full notes). What changed since 0.7.0:

• 0.17.1: the AMD engine can be tried on a card setup calls unsupported (amd_setup.py --ignore-old-graphics, at your own risk).
• 0.17.0: an Apple Silicon Mac transcribes on its own GPU, about twice the CPU's speed, on large-v3-turbo; a video in another language is left alone before it downloads.
• 0.16.0: the server runs Kitsune-Transcribe models next to Whisper: small Japanese-only recognisers distilled from Cohere Transcribe. Pick kitsune-0.6b or -0.1b at setup, any size in the popup, in bf16, fp16, int8, fp8, NVFP4 or MXFP4; setup installs PyTorch for them.
• 0.15.0: the server can run Whisper on an AMD graphics card (experimental); setup offers it where it finds a Radeon RX 7000/9000 or Ryzen AI GPU (RX 6800/6900 on Linux too). Setup also remembers the browser whose YouTube cookies every server start sends, for YouTube's "Sign in to confirm you're not a bot".
• 0.14.6: mining finds the fields in any case (Eminent's picture, sentenceAudio), and a card without them names its own. With "Particles count as known", the copula is found where Firefox's splitter hid it (大変だったね) and everyday grammar words (ある, する, この, こと …) count as known; a card for one decides its colour.
• 0.14.5: word colours show only what your Anki deck says. Particles (は, です) and names and Latin text (OK) are no longer coloured without a card; both are switches, off by default, in a new "More word colour options" drawer with the katakana switch and your known-words list.
• 0.14.4: speech Whisper skipped at the start of a stretch of audio (up to 20 seconds with no subtitle, even though someone was talking) is transcribed again and filled in. Videos you watched before are transcribed again on your next visit.
• 0.14.3: Alt+Shift+H hides the status badge on the video, the red "server offline" too (a popup switch as well).
• 0.14.0: your own list of known words (Alt+Shift+K adds the word under the pointer); particles, katakana words, names and Latin text can be coloured; a card for a noun also colours its する forms; the pitch overbar reads Yomitan's pitch graph.
• 0.10 to 0.13: word colours by each word's Anki card state and a pitch accent overbar, both off by default; setup asks which Whisper model to download; only the tab you are watching is transcribed; subtitles pause while the speech is not Japanese; sung lyrics get subtitles; lines are whole sentences, never cut inside a word; a mined card gets the line on screen; Left steps back a line; the popup offers to update the server, and tells you when a newer release is out.
• Also: live streams; a master switch (Alt+Shift+S); the Whisper model picked in the popup; any installed font for the subtitles; a new card matched to its subtitle by sentence and word, with the screenshot and audio prepared while the line plays; Chrome support.

0.7.0 was the last version on this listing; 0.8.0 to 0.14.0 came out on GitHub only (https://github.com/Multysquid/shisu-ko/releases has every release's full notes). What changed since 0.7.0:

• 0.15.0: the server can run Whisper on an AMD graphics card, experimental and not yet tested on AMD hardware by the maintainer. Setup offers it where it finds one: Radeon RX 7000/9000 and their PRO cards, Ryzen AI 300 (890M/880M) or Ryzen AI Max graphics on Windows (Adrenalin 26.2.2 or newer), these and RX 6800/6900 on Linux with ROCm 7.2. Setup (or --save-cookies-from-browser) also remembers the browser whose YouTube cookies every server start sends, the Start button's too, for YouTube's "Sign in to confirm you're not a bot".
• 0.14.6: mining finds the fields in any case (Eminent's picture, sentenceAudio), and a card without them names its own. With "Particles count as known", the copula is found where Firefox's splitter hid it (大変だったね) and everyday grammar words (ある, する, この, こと …) count as known; a card for one decides its colour.
• 0.14.5: word colours show only what your Anki deck says. Particles (は, です) and names and Latin text (OK) are no longer coloured without a card; both are switches, off by default, in a new "More word colour options" drawer with the katakana switch and your known-words list.
• 0.14.4: speech Whisper skipped at the start of a stretch of audio (up to 20 seconds with no subtitle, even though someone was talking) is transcribed again and filled in. Videos you watched before are transcribed again on your next visit.
• 0.14.3: Alt+Shift+H hides the status badge on the video, the red "server offline" too (a popup switch as well). Every GitHub release carries an .xpi, the unsigned build while addons.mozilla.org reviews it; this listing gets selected releases, each as its number plus .1.
• 0.14.1: submitted to this listing with everything below.
• 0.14.0: your own list of known words (Alt+Shift+K adds the word under the pointer); particles, katakana words, names and Latin text can be coloured; a card for a noun also colours its する forms; the pitch overbar reads Yomitan's pitch graph.
• 0.10 to 0.13: word colours by the state of each word's Anki card and a pitch accent overbar, both off by default; setup asks which Whisper model to download; only the tab you are watching is transcribed; subtitles pause while the speech is not Japanese; sung lyrics get subtitles; lines are whole sentences, never cut inside a word; a mined card gets the line that was on screen; Left steps back a whole line.
• 0.9.0: the popup offers to update the server (the new notifications permission is for that); 0.8.0: a Start server button (optional nativeMessaging permission).
• Also: live streams; a master switch (Alt+Shift+S); the Whisper model picked in the popup; any installed font for the subtitles; a new card matched to its subtitle by sentence and word, with the screenshot and audio prepared while the line plays; Chrome support.

0.7.0 was the last version on this listing; 0.8.0 to 0.14.0 came out on GitHub only (https://github.com/Multysquid/shisu-ko/releases has every release's full notes). What changed since 0.7.0:

• 0.14.5: word colours show only what your Anki deck says. Particles (は, です) and names and Latin text (OK) are no longer coloured without a card; both are switches, off by default, in a new "More word colour options" drawer with the katakana switch and your known-words list.
• 0.14.4: speech Whisper skipped at the start of a stretch of audio (up to 20 seconds with no subtitle, even though someone was talking) is transcribed again and filled in. Videos you watched before are transcribed again on your next visit.
• 0.14.3 (listed as 0.14.3.1; 0.14.2 had no release of its own): Alt+Shift+H hides the status badge on the video, the red "server offline" too, and shows it again (a popup switch as well). Every release on GitHub carries an .xpi: signed by addons.mozilla.org within minutes or, while it reviews a version, the unsigned build until the signed one replaces it. This listing gets selected releases, each as its number plus .1.
• 0.14.1: submitted to this listing with everything below (0.14.3.1 takes its place if its review had not finished).
• 0.14.0: Word colours: particles count as known and show green (a popup switch, on by default); names and Latin text are blue; your own list of known words, and Alt+Shift+K puts the word under the pointer on it; katakana words can count as known; a card for a noun also colours its する forms; the pitch overbar reads Yomitan's pitch graph.
• 0.13.0: a word mined in kanji is coloured where the subtitle writes it in kana (更に in さらに); a long line is never cut inside a word.
• 0.12.0: lines that are sentences, with their punctuation, never two sentences on one row; real lines are no longer thrown away as noise.
• 0.11: sung lyrics get subtitles; "No speech found in this video" when nothing was heard; a mined card gets the line that was on screen; lines merge across Whisper's segments and break between words; Left steps back a whole line.
• 0.11.0: word colours by the state of each word's Anki card (green, yellow, orange, red) and a pitch accent overbar, both off by default; setup asks which Whisper model to download.
• 0.10: only the tab you are watching is transcribed; subtitles pause when the speech is not Japanese.
• 0.9.0: the popup offers to update the server (the new notifications permission is for that); 0.8.0: a Start server button (optional nativeMessaging permission).
• Also: live streams; a master switch (Alt+Shift+S); the Whisper model picked in the popup; any installed font for the subtitles; a new card matched to its subtitle by sentence and word, with the screenshot and audio prepared while the line plays; Chrome support.

0.7.0 was the last version on this listing; 0.8.0 to 0.14.0 came out on GitHub only (https://github.com/Multysquid/shisu-ko/releases has every release's full notes). What changed since 0.7.0:

• 0.14.1: the first version on this listing since 0.7.0, with everything below.
• 0.14.0: Word colours: particles count as known and show green (a popup switch, on by default); names and Latin text are blue; your own list of known words, and Alt+Shift+K puts the word under the pointer on it; katakana words can count as known; a card for a noun also colours its する forms; the pitch overbar reads Yomitan's pitch graph.
• 0.13.0: a word mined in kanji is coloured where the subtitle writes it in kana (更に in さらに); a long line is never cut inside a word.
• 0.12.0: lines that are sentences, with their punctuation, never two sentences on one row; real lines are no longer thrown away as noise.
• 0.11: sung lyrics get subtitles; "No speech found in this video" when nothing was heard; a mined card gets the line that was on screen; lines merge across Whisper's segments and break between words; Left steps back a whole line.
• 0.11.0: word colours by the state of each word's Anki card (green, yellow, orange, red) and a pitch accent overbar, both off by default; setup asks which Whisper model to download.
• 0.10: only the tab you are watching is transcribed; subtitles pause when the speech is not Japanese.
• 0.9.0: the popup offers to update the server (the new notifications permission is for that); 0.8.0: a Start server button (optional nativeMessaging permission).
• Also: live streams; a master switch (Alt+Shift+S); the Whisper model picked in the popup; any installed font for the subtitles; a new card matched to its subtitle by sentence and word, with the screenshot and audio prepared while the line plays; Chrome support.

Model picker: the popup's "Transcription model" field takes any faster-whisper size or Hugging Face CTranslate2 repo id; sent with every /sync; the server validates the name (never a path), canonicalises aliases, downloads on a side thread while the old model keeps subtitling, swaps between windows, restarts sessions, keeps the other model's cue cache as <id>.<model>.cues.json, reports model/model_loading/model_error (+ /health lists downloaded models). Overlay shows "Loading model X…" / the error.
Font family: any installed font in front of the preset stack, with a live preview and an installed/not-found probe in the popup; sanitised before CSS.
Launcher update step (server/update.py): run.cmd/run.sh fast-forward a git checkout or fetch the newest release zip first; --no-update skips it.
Stale .part downloads that YouTube answers with 416 are discarded and downloaded afresh.
AMO listing package under docs/amo/ + publish-addon.cmd; homepage_url in the manifest; .amo-upload-uuid untracked.
Docs: README/AGENTS.md sections for all of the above, live streams marked Firefox-only, README/reviewer notes/privacy policy aligned with what 0.6.0 actually sends.
Tests: 290 server (from 190), 136 addon (from 73).