Changelog

What changed in each version of BelliedMonkey Translator, newest first. The same text goes to the App Store, Chrome Web Store and Firefox Add-ons; the source of every release is on GitHub Releases.

1.9.0 2026-09-10

• New free allowance: sign in to claim a small allowance we provide — enough to translate a few hundred pages, usable for translation, read-aloud, and transcription. When it runs out, add your own API key or join the community. Prefer not to sign in? Nothing changes — bring your own key or use the free Google engine.
• Fix: on the setup screen, tapping “Continue” right after pasting a key could drop the key in some cases; it now saves it before moving on.
• Secondary text in light mode is a touch more legible.

1.8.0 2026-09-07

• Two new things:
• Podcasts and videos with no subtitles can now get bilingual subtitles. Tap "AI transcript subtitles" where it says subtitles are unavailable, and the audio goes to the transcription engine you configured (live streams show word-by-word on the video, with a full-sentence history beside it).
• New in the app: "Conversation · live interpreter". They speak, you read Chinese; hold "I'll talk" to speak Chinese and show the translation — read aloud too. Keeps listening while locked; finished sentences go into review.
• The app's home screen is redesigned around one card: what's due today.
• Privacy: audio goes only to the transcription endpoint you configured — our servers never touch it. No recording is kept, only text.

1.7.18 2026-09-07

• Three fixes:
• In the app, once you set up a speech engine, Podcast Mode and the card's "Listen" button now appear right away — no relaunch needed.
• The app's review screen no longer shows a "Continue in the app" button (that one belongs to the browser extension).
• Links and light grey text that were hard to read in Dark Mode (for example "Get a key" in Settings) are now legible; light-mode buttons and secondary text are a shade darker too.
• The 1.7.17 fix is still here: your API key no longer disappears after one-tap setup.

1.7.17 2026-09-06

• Fix: in the app's Settings, "One key sets up all three" lost the API key — and the three
• test rows — the moment you tapped the button. The key now stays put and each test reports
• on its own line.
• Everything from 1.7.16 is still here: the redesigned sign-in (Apple / Google first),
• feedback and rating entry points on every surface, and anonymous usage data with a single
• switch in Settings to turn it off.

1.7.16 2026-09-05

• Sign-in, redesigned: Apple and Google one-tap sign-in sit at the top; email is folded into
• a single line and expands only when you need it. The code step now replaces the form
• instead of stacking under it.
• Feedback and rating are one tap away on every surface — settings, the popup, the end of
• onboarding, and the app's settings — to write to the developer or rate on the store.
• Anonymous usage data is now sent: only which features are used, on which browser, and
• whether a translation succeeded or failed — never the pages you read, the text, the
• addresses, your keys or your account. On by default, one switch in Settings turns it off,
• and turning it off deletes what this device sent. Privacy policy section 5 lists every field.

1.7.14 2026-09-05

• Settings in the app now match the browser extension.
• The Quick / Detailed split is in the app. One key used to configure translation,
• speech and transcription in the extension — but the app still made you fill in
• every field by hand. Same setup on every device now.
• Engines that need no key no longer show a key field. Picking Google Translate and
• still being asked for a secret was confusing, and it was there until now.
• More interface languages, so Hindi and others no longer fall back to English.
• Speech now needs you to pick an engine.
• It used to default to the system voice. That voice isn't good enough to learn from,
• and nobody ever chose it. Now nothing is read aloud until you configure an engine —
• one step, under Settings › Speech.
• This also fixes an embarrassing one: it said "not configured" and yet the preview
• button really did speak in the system voice, while the screen claimed "playing".

1.7.12 2026-09-03

• The big one: **no password to remember, no code to wait for.**
• Sign in with Apple or Google, in one tap. Until now the only way in was an emailed
• six-digit code — leave the app, find the mail, copy the digits back. Email still
• works if you would rather not use a third-party account.
• Rebuilt first-run setup. One API key now configures translation, read-aloud and
• transcription together, and it tests itself right there — you find out it works
• before your first page, not after.
• Sentences you actually stop and finish reading turn into review cards, and capture
• is on from the start (scrolling past does not count).
• If you sign in as a different account while this device still holds another
• account's material, it now says so — nothing was lost; sign back in and it returns.
• Keyboard shortcuts and inline code in a paragraph (Super + Return, for example) no
• longer vanish from the translation.

1.7.5 2026-09-03

• This release finally connects the whole path: set it up, translate a page, keep what you read.
• Rebuilt first-run setup. One API key now configures translation, read-aloud and
• transcription together, and it tests itself right there — so you find out it works
• before you translate your first page, not after.
• Sentences you actually stop and finish reading turn into review cards, and capture
• is on from the start (scrolling past does not count). It used to be off by default,
• so people read a whole page and found nothing saved.
• After translating on the web you can go straight back into the app to review. If the
• two sides are signed in as different people, it now says so instead of quietly
• failing to line up.
• Keyboard shortcuts and inline code in a paragraph (Super + Return, for example) no
• longer vanish from the translation. They used to be dropped entirely, leaving a
• sentence that read fine but said the wrong thing.
• Podcast mode reads cards whose language could not be identified. On iPhone, those
• cards previously would not play at all.

1.6.8 2026-08-29

• The first launch now walks you through actually getting translation working: pick
• your languages, set up an engine, enable the Safari extension, then translate a
• page right there to see it work.
• You can look around without signing in. An account is only needed when you want
• review progress synced to another device.
• After installing on iPhone, there's a new page at belliedmonkey.cc/setup.html:
• follow the three steps to enable the extension in Settings, come back, and it
• turns green by itself to confirm it worked — a question that previously had no
• answer anywhere.
• Two places that used to leave you waiting are fixed: a failed translation now
• says so and offers a retry you can tap, instead of sitting at "Translating…"
• forever; and an API Key you type in Settings is no longer lost if you lock the
• screen or switch apps before leaving the field.

1.6.7 2026-08-26

• Podcast mode now works like a music app:
• Playback keeps going when you leave the app, lock the phone, or switch away —
• card after card, without stopping.
• The Lock Screen shows the card being read: the sentence, its translation, and —
• while the analysis plays — three lines (vocabulary / phrases / grammar) that
• light up one at a time as they are spoken.
• The Dynamic Island shows your progress; press and hold to see the full sentence
• and translation.
• Play/pause and next work from the Lock Screen, your car and your headphones.
• "Previous" repeats the current card from the top.
• Playback resumes by itself after a phone call ends.
• Also: the screen no longer dims while podcast mode is in front. To keep seeing the
• card after you lock the phone, turn on Always-On Display (Settings → Display &
• Brightness → Always On).
• Note: this version changes how the analysis is split for speech, so previously
• downloaded analysis audio needs to be preloaded once more for fully offline playback.

1.6.6 2026-08-24

• Fixed: on React-rendered sites such as Substack, turning translation on made text
• impossible to select and hyperlinks impossible to open.
• The translation was being inserted as a sibling inside a container the page's own
• framework owns, so every change of selection made the framework re-lay-out every
• paragraph in the article — which wiped the selection, and moved the paragraph out
• from under the pointer between mouse-down and mouse-up so the click never reached
• the link. The translation now goes inside its own paragraph and leaves the page's
• structure alone.
• Also:
• When the speech endpoint cannot be reached it now says so and stops within 20
• seconds, instead of waiting for the system timeout.
• Stability fixes in capture and review.
• (The podcast preload feature is part of the macOS / iOS app, not the extension.)

1.6.5 2026-08-23

• Podcast mode can now be taken offline. Tap Preload before you set out and the speech,
• sentence notes and translations for today's deck are downloaded to your device, so a
• whole session plays with no network at all.
• The first tap only prices it — it shows how many cards and how many billed calls, and
• spends nothing. You decide, then tap again.
• Cards start faster. Everything a card needs is now requested at once instead of one
• piece at a time, so the silence between sentences is gone.
• Cards captured without a translation can be given one, using the engine you already
• configured.
• When the speech engine cannot be reached, it now says so and stops — instead of
• skipping through the whole deck in silence and reporting that the round is finished.
• Stop now takes effect promptly during a preload.

1.6.4 2026-08-22

• Reasoning models now translate long passages reliably. Some of them used to return a
• successful response with no text in it — the paragraph just stayed blank, with no error
• to explain why. Requests now carry only the fields each endpoint is known to accept,
• which fixed that and cut a long paragraph from about 15 seconds to 3.
• Advanced settings accept your own request parameters, so a private or corporate
• endpoint can be tuned without waiting for us to add it.
• More engines to choose from — each one measured against its real endpoint rather
• than its documentation.
• An engine whose default model had been retired no longer fails the moment you pick it.
• Fixed a few interface strings that showed their formatting marks verbatim.

1.5.3 2026-08-19

• Custom endpoints are now used exactly as you type them, with no exceptions. Failed
• requests quote the server's own explanation instead of a generic message, and the
• request body concedes any optional field the server says it will not accept. The
• translation cache is keyed by endpoint and model, so changing either takes effect
• immediately, and 「Test connection」 always makes a real request.

1.5.1 2026-08-15

• 1.5.1 — Meet the BelliedMonkey
• A full visual redesign:
• New mascot icon and in-page floating button — a pot-bellied monkey carrying the translation
• Warm terracotta & sage palette across every surface (popup, settings, review, app)
• Translations default to a calmer sage green; your custom color is untouched, old defaults migrate once automatically
• Translation, learning and privacy model unchanged

1.5.0 2026-08-13

• 1.5.0 — Listen, speak, read, write: every sentence, every skill
• The learning module now trains all four skills:
• New exercises: translation multiple-choice, blind-listen word picking, read-aloud scoring, AI comprehension questions
• Each review automatically rotates to your least-recently-verified skill — a sentence counts as learned only while all four stay fresh
• Read-aloud scoring: your recording goes only to the transcription endpoint you configure, and is discarded once the transcript returns — it never touches a server of ours. No endpoint configured, no speaking exercise.
• Optional AI-generated questions use your own API key — one call per card, cached forever
• Every new exercise has a zero-cost local variant — fully usable with no API configured at all