What's New
v2.1.0
- Voiceover falls behind the conversation less: when translated phrases pile up, they are voiced a little faster, with the voice's pitch unchanged — there is nothing to set up. With MeetVoice Desktop 2.1.0, the pauses before and after each phrase are also shorter with the main voice and studio voices.
- While the translation is voiced into the call, your own voice is now turned down to 5% instead of 20%, so the other side hears the translation more clearly. The “Original voice during translation” slider now goes down to 5% in steps of 5, and a setting of 20% moves to 5% automatically. If you don't want your own voice in the call at all, a hint under the slider points to the “Only the translation goes to the call” switch. An older version of the extension on another computer with the same Chrome profile treats 5% as 10%.
- Changes to the “Translated voice” and “Original voice during translation” sliders made before you start translating now reach an open Meet tab right away, without reloading it.
- Voice previews now say a line from a meeting instead of a greeting, so you hear how a voice sounds in a conversation; for voices other than studio ones, this needs MeetVoice Desktop 2.1.0.
- Smaller fixes: screen readers no longer start reading the subtitles over again each time they update while someone is speaking, and in MeetVoice Desktop, “Disconnect license on this computer” now works when the activation was already removed on the license server's side, instead of failing with an error on every click.
v2.0.9
- Pro now costs €19 a month. New subscriptions are billed monthly only; existing yearly subscriptions keep working.
- New Max plan, €39 a month: 30 hours of translation and studio voices for the voiceover in both directions, in 27 languages. Studio voices need MeetVoice Desktop 2.0.9 or later.
- On Max, you choose a studio voice in the extension and can preview it; the extension and MeetVoice Desktop show which voice is speaking and, when the standard voice takes over, why.
- Voiceover is more dependable (with MeetVoice Desktop 2.0.9): since about October 10 it could go silent when the main voice was unavailable, because the backup voice had stopped working — it works again. If the main voice is unavailable when a call starts, it now comes back as soon as it recovers instead of leaving the backup voice until the end of the call, and voiceover no longer switches to the backup voice toward the end of the month.
- A brief drop in speech recognition (up to 10 seconds) no longer shows the “Server disconnected” banner: translation picks up on its own, and with “Only the translation goes to the call” turned on, the other side hears your own voice in the meantime.
- Smaller fixes in MeetVoice Desktop: turning translation back on right after turning it off no longer leaves it off, and when another program is using the app's network port, MeetVoice says so instead of reporting repeated crashes.
v2.0.8
- Translation hours on paid plans now renew every month on the day and at the time you subscribed, instead of on the 1st of each calendar month; yearly plans get their monthly hours on the same schedule. Subscriptions bought before October 8, 2026 switch to the new schedule at the first monthly renewal date for their hours on or after October 8; until then, their hours keep resetting on the 1st of the month.
- MeetVoice Desktop shows the exact date and time your hours renew, in your local time.
- Desktop app: updated Electron engine with the latest Chromium security fixes.
v2.0.7
- The glossary is easier to fill in. “Do not translate” is now a list of names: type a name and press Enter, remove it with the cross. “Translate as” is a list of rows with a term and its translation, so you no longer type the arrow yourself; you can also paste two columns from a spreadsheet or lines such as “term → translation” (→, =>, ->, =, a tab or a spaced dash all work as the separator). Messages about the limit, a repeated entry or a missing translation appear next to the field, a failed save is reported instead of passing silently, and a pair written as “a => b” is no longer saved incorrectly. Area chips such as “Finance” or “Legal” now toggle — click one again to remove it — and a glossary edit made just before you start translating is included in that session.
- Select a word or phrase in the transcript and an “Explain” button appears right at the selection; the explanation opens in a card at the bottom of the panel. The card offers “Do not translate”, which adds the term to your names, and “Set translation”, a small form that adds or updates a “Translate as” pair. Selecting text in the transcript now also works on Google Meet pages, where Meet's own styles used to block it and “Explain” could only show a hint.
- The transcript panel shows what is going on: a loading state, “Phrases will appear here during translation” while there is nothing yet, and “Couldn't load history” with a “Retry” button if loading fails. The panel has a title and a label for screen readers, Esc closes the card first and then the panel, the panel's notices appear inside it, and MeetVoice error notices on the Meet page no longer end up under the open panel.
- Speech recognition is more dependable (with MeetVoice Desktop 2.0.7): when translation starts or recognition reconnects, the last second and a half of what you said while it was connecting is kept instead of being lost; a connection to recognition that hangs while opening is dropped after about ten seconds and retried automatically, instead of leaving translation silent until you press Stop; and a long mute in Google Meet no longer makes the app reconnect recognition every minute, which could end outgoing translation with an error — recognition now resumes when you unmute.
- Voiceover: muting it in the extension window or with Alt+M now also works while a session is still connecting, and it stays muted if you then close the extension window. A voice gender change made while connecting reaches the session too; the voice choice and the voiceover mode are locked for those few seconds and unlock once the session connects or fails to start. A network hiccup while a phrase is being voiced no longer freezes the voiceover — within about five seconds the phrase is voiced by the backup voice or skipped — a slow voice service no longer adds about ten seconds to every phrase, and stopping a voice preview while it is still loading clears the spinner.
- Long translated phrases spoken into the call are no longer cut off after 15 seconds.
- Speaker names stay correct in long meetings: the audio sent to speech recognition no longer slowly runs ahead of the call's audio.
- Japanese, Korean and Chinese text now wraps between words and phrases instead of mid-word in subtitles, the transcript panel and History, and Chinese in subtitles and the panel is shown in a Chinese font instead of a mix of Japanese and Chinese character forms. History uses the right plural forms for the number of entries, shows durations in your interface language and no longer shows a count for an empty meeting. Serbian dates and durations in the extension use Latin script, and the Korean interface no longer addresses you as «당신» in status lines.
- In Markdown and Notion exports, participant names and transcript text are escaped, so a name such as “[Pay](https://…)” stays plain text instead of turning into a link.
- If MeetVoice Desktop is closed or unreachable for several minutes during a call, translation now stops with a clear message that the connection to the server was lost, and you can start it again as soon as the app is back, without closing the Meet tab. After you reload the Meet tab during a session, the transcript panel shows the earlier lines again, and a meeting whose language pair you changed during the call keeps its final pair in History, exports and summaries however it ends. When speech recognition is lost, the disconnected indicator and the error notice now stay on screen, and Alt+S and the volume sliders respond right after the browser wakes the extension.
- Desktop app: installing an update is more dependable. A double click on “Restart to update” no longer opens two confirmations or starts two installs, translation cannot be turned on while the app is closing for the update, and if an install does not finish, MeetVoice comes back in a working state and shows the update error. You will notice this starting with the update to 2.0.7: an update is installed by the version you are updating from, and these fixes are already in 2.0.6. A downloaded update now stays ready: later checks, even offline ones, no longer discard it or announce it again. Closing the license window or choosing the free trial while a key is being activated no longer locks you out, a late reply from the license server can no longer erase a key you have just activated, and restarting translation from the desktop window right after starting it no longer leaves it stopped.
- Desktop app: for extra protection against tampering, it no longer reads the NODE_OPTIONS and NODE_EXTRA_CA_CERTS environment variables. If your company network relied on them to trust its own security certificate, MeetVoice may be unable to connect there.
- Smaller fixes: in multilingual meetings, a short word that recognition cannot assign to a language (such as “okay” between German and Russian) no longer leaves the phrase before it untranslated and unvoiced; a long glossary with names in scripts such as Thai, Arabic, Hebrew or Bengali no longer stops speech recognition from starting; and a phrase that ends with a number, a time or an amount no longer waits five to six seconds for its last words (in English, such numbers may now appear spelled out).
v2.0.6
- The option to use your own Deepgram key has been removed: speech recognition always runs on your plan or trial. Multilingual meetings now work on any plan and on the free trial, with no key of your own, and their time counts toward your plan hours. Multilingual mode offers English, German, Russian, Spanish, Portuguese, French, Italian, Japanese, Dutch and Hindi; other languages are no longer available in it. A key you saved earlier is deleted from the extension.
- New “Only the translation goes to the call” option under “What others hear” in the sound settings: the other side hears only the translated voice, not your own. If translation stops working — you press Stop, the connection drops or the server reports an error — your own voice returns to the call automatically. A connection that silently dies, for example after a network change, is now noticed within about 15 seconds and reconnects (with MeetVoice Desktop 2.0.6).
- The translated voice catches up instead of piling up: a phrase whose voice-over has been waiting to play for more than about 10 seconds is skipped, while its subtitle still appears, and a stale translation of your own speech is no longer spoken into the call.
- What other participants say now appears with punctuation and capital letters in subtitles, the transcript and exported files.
- A clearer glossary: the fields are now called “Meeting topic”, “Do not translate” and “Translate as”, each with a hint; area chips such as Finance or Legal replace the Tech/Finance buttons; the pairs field shows an example in your own languages, counts pairs and says why a line was skipped. Ordinary words added by the old Tech/Finance buttons are translated again, suggestions of words the recognizer was unsure about have been removed, and edits are no longer lost when the extension window closes, including with Esc.
- Fixed: “Translate as” pairs now really work in both directions, as promised in 2.0.5: a “deal → сделка” pair also turns “сделка” in your speech into “deal”, and your own reverse pair takes priority. A long glossary in Cyrillic, Hindi or Japanese no longer stops subtitles from appearing.
- The desktop app is easier to follow: its window shows one clear next step, the language pair and duration of the current translation, and when your hours reset. Errors explain the cause and offer a fix and a “Copy diagnostics” button, buttons report failed actions instead of staying silent, and notifications open the matching card. Setup counts as complete once the first subtitles appear in Google Meet.
- When your trial ends or your plan is no longer active, MeetVoice stays in the tray (the menu bar on Mac) and shows how to continue instead of closing, and the extension names the reason right away instead of a long wait and a generic error. The tray menu is shorter; the license window has a “Paste” button and a “Lost your key?” link and no longer covers the checkout page; subscribers change their plan in the billing portal.
- Settings → Advanced has a new switch for usage statistics: daily counts of which features and settings you use — no speech, transcripts or names — sent through the desktop app to help us decide what to improve. On by default; turning it off stops collection.
- Desktop app: updated Electron engine with current Chromium security fixes, the app takes less disk space, and a license agreement is shown on first install. macOS 13 Ventura or later is now required; on older macOS the app stays on version 2.0.5.
- Smaller fixes: consistent icons instead of emoji across the extension; a new installation picks a language pair from your browser language; several Google Meet tabs no longer mix up transcripts, names or voice-over; no false “Server disconnected” banner after Stop; speaker names come back when Meet rebuilds the participant grid; moving the “Original voice during translation” slider no longer makes you quieter before the translated voice actually starts; the “Test conversation” ends with an error at once when speech recognition fails for good instead of running out the clock; deleting a meeting also deletes its AI summary; Serbian dates and numbers in the desktop app use Latin script; the extension no longer asks for access to api.deepgram.com.
v2.0.5
- The glossary is more flexible: add a meeting topic, important names and exact “spoken → translated” pairs. MeetVoice uses them in both translation directions so terms and product names stay consistent.
- Glossary changes now reach the active translation session reliably, and invalid saved data no longer breaks the settings panel.
- Demo and voiceover diagnostics are more precise, making it easier to confirm that translation and speech are ready before a call.
v2.0.4
- People on speakers no longer hear the translation of their own words: the extension now explicitly requires Chrome’s echo cancellation, so our own voice-over stays out of the microphone.
- The Google Meet control panel has been reworked: one compact panel with a drag handle, collapsing and toggles for transcript, subtitles and translation audio. Its position is remembered, and “Stop translation” lives in a separate menu and never ends the call.
- New in the transcript: select a word or phrase and choose “Explain selected term” to get an explanation in the context of that line, then add it to your glossary in one click. Requires the desktop app running and an active plan.
- Cleaner audio and subtitles: the original is quieted only while the translated voice is actually speaking and comes back in its pauses, and subtitle lines wrap to the real font width, stop jumping while the translation is still streaming and no longer slide off-screen when dragged.
- Speech no longer splits numbers and dates in the middle, one-word answers are no longer silently skipped, and closing quotation marks are read together with the sentence in every language.
- More reliable stop and restart: stopping works even when saving the meeting stalls, starting right after a stop no longer shows a false error, and reconnecting no longer drags leftovers from the previous session along.
v2.0.3
- Control translation from inside Google Meet — a bar with the session status and buttons for the transcript, the microphone and stopping, so you no longer have to open the extension window mid-call.
- Reworked extension window: clearer setting names, all volumes grouped under “Sound settings” with a reset button, and a line telling you whether you are ready to start and what is missing.
- Deleting a meeting from the history can now be undone.
- The pre-call check is honest now: it tells you which language you will hear the translation in, notices a muted microphone instead of showing a green tick, and counts as passed only after a full round of translation.
- A simpler first run of the desktop app — an overview screen, a step-by-step setup guide and the status of its connection to the extension.
- Calls no longer end unexpectedly: quitting the app or disconnecting a license during translation asks for confirmation, and activating a license applies to the next session instead of cutting the current one.
- Smaller fixes: the extension no longer asks for access to translate.meetvoice.app, and a false start error no longer appears after “Stop”.
v2.0.2
- Microphone in Google Meet works again — Meet had stopped transmitting your voice when the extension was installed, while still showing the mic as live.
- Eight new languages: Greek, Swedish, Norwegian, Finnish, Danish, Hindi, Chinese and Vietnamese — 26 in total.
- Transcript in its own floating window — pop the panel out and keep reading the translation while you work in another window.
- Separate voices per direction — pick one voice gender for what you hear and another for how you are voiced to the other side.
- Voiceover keeps up with the conversation — when a speaker moves on, the outdated voiceover is dropped instead of playing over the new sentence.
- Meeting summary and transcript export work again — they were rejected by the desktop app in 2.0.1.
- Clearer prompts when the desktop app is out of date: “update”, not “restart”.
v2.0.1
- Test conversation with AI: try live translation in a two-minute practice call right from the extension — no meeting or second participant needed, and a results card confirms your setup works.
- New HD voices: spoken translation now uses high-definition voices by default — they start speaking sooner and sound more natural. HD voices carry a badge in the voice picker, and if they are ever unavailable, the previous voices take over automatically.
- More reliable session start: the Start button now waits until translation is fully ready and shows clear progress — no more sessions that seem to start but stay silent.
- Voiceover recovers on its own: if spoken translation stumbles, it retries automatically, and the extension lets you know when quality degrades and when it is back to normal.
- Fewer false license errors: brief network or payment-service outages are no longer mistaken for an invalid license.
- Polish throughout: simpler wording across the apps, faster translation start, clearer error messages for your own speech-recognition key, plus stability and security updates.
v2.0.0
- Speech recognition included by default: every plan and the free trial now include speech recognition, so there's no API key to set up. Bringing your own Deepgram key remains available as an optional, more private choice.
- New subscription plans: choose Starter, Pro, or Business, each available with monthly or yearly billing.
- Simpler settings: the translation engine is now chosen automatically, and the extra API-key fields for it have been removed.
- macOS app built for Apple Silicon: a native version for M1 and newer Macs.
- Steadier long meetings: improved audio flow control keeps translation running smoothly during extended calls.
v1.6.1
- Fixed macOS update installs: downloaded DMGs no longer show "disk image damaged" when you open them.
- More reliable multilingual meetings with your own API keys: multilingual sessions now use the same translation fallback as single-language mode, and subtitles are less likely to drop after language switches.
- Smoother first-time setup: the welcome tab no longer reopens every time you open the extension popup, and setup progress and final hints are fully localized in all supported languages.
- Trial starts correctly from the desktop app: choosing "Try for free" on a fresh install now properly begins your trial and starts the local server.
- More reliable session start: starting translation from the popup works better when the desktop connection URL is still being discovered.
v1.6.0
- Live Edge voice updates: changing the voice now applies during the meeting without restarting the session, so you can adjust the sound more naturally.
- Safer extension settings: Deepgram keys, glossary entries, subtitle preferences, and connection options are saved more carefully and are less likely to reset or break a working setup.
- More reliable transcript exports: downloads handle edge cases more safely, so long meetings and unusual transcript content are less likely to produce broken files.
- Clearer desktop update experience: the app explains update status and next steps more plainly, making it easier to know when MeetVoice is ready.
- Sturdier background work: the local server, translation proxy, and website analytics checks now handle startup, shutdown, and failed requests more gracefully.
v1.5.9
- More accurate multilingual meetings: when participants speak different languages, MeetVoice now recognizes each person's language more accurately and mixes languages up less in the subtitles — even when someone switches language mid-sentence. Fewer mistranslated lines and faster, steadier subtitles.
- Steadier language detection: short replies and the occasional foreign word are handled more accurately, so the subtitle language no longer flickers on individual words.
- Simpler connection settings: the local server URL is hidden by default, manual mode is available only when needed, and there is less risk of accidentally breaking settings that already work.
- Helpful popup tooltips: short explanations near multilingual mode clarify when to choose one primary language, several languages, or auto-detection.
v1.5.8
- The glossary is smarter: MeetVoice now suggests words and names that speech recognition was not fully sure about, so you can quickly add them to the glossary and improve accuracy in the next phrases.
- Auto-added participant names are easier to understand: the extension window shows how many names are already used for the current meeting, and the hint appears only during an active session.
- New quick presets for work meetings: add basic terms for Tech and Finance with one click — less manual setup before a call.
v1.5.7
- Status dashboard in the desktop app (Windows): open it from the tray icon to see whether the server is running, check your license or trial, manage updates, and jump to the website, Chrome extension, or Deepgram console.
- First-time setup checklist in the desktop app: at a glance you can see whether the extension is connected, a Deepgram key is saved, and you have opened a Google Meet call.
- Flexible updates on Windows: choose semi-automatic update checks or manual-only — fewer surprise restarts during an active meeting.
v1.5.6
- Separate voices for incoming and outgoing translation. Settings now has two independent voice pickers: one for what you hear (the voice translating your counterpart), one for what meeting participants hear from you. Configured per language.
- Improved speaker name detection. Several bugs fixed where speaker names in the transcript appeared incorrectly or were missing. Names now appear faster and more accurately, especially for short utterances and at the start of a call.
- Better translation for similar language pairs. Translation quality improved for Ukrainian↔Russian, Portuguese↔Spanish, Dutch↔German and other close pairs — fewer literal translations, more natural phrasing.
- API keys stored in the OS secure vault. Deepgram and other keys are now encrypted via the Windows/macOS system keychain instead of being stored in plain text.
- Transcript button moved to the right side of the screen — no longer overlaps Google Meet controls.
v1.5.5
- Mixed-script subtitles: when a phrase in the source language (for example, English) includes names in another script — "Alexandr and Serhiy will join..." — the translation is no longer dropped due to a false SKIP. Previously such lines sometimes never appeared on screen even though they still needed to be translated.
- Stronger protection for the automatic name glossary from Meet: harmful or accidental snippets such as "System:" / "User:" in a participant's display name are filtered further before they reach the translator, so one person's renamed label is less likely to affect subtitles for the whole meeting.
v1.5.4
- Participant names are now preserved verbatim across translation: if your colleague's name is "Maciej", they stay "Maciej" in the Russian transcript instead of becoming "Мацей". This applies to all 18 languages — Latin, Cyrillic and CJK scripts are now passed through letter-for-letter via the automatic glossary. Searching for a colleague by name in subtitles, meeting history and SRT/TXT exports now works the same on both sides of the call.
- Subtitle stability in real time: the extension now correctly translates speech fragments and disfluencies ("so, like, basically...", "um, the thing is..."), no longer treating them as technical noise. Previously such fragments would sometimes disappear from the overlay or get padded with words nobody actually said. Especially noticeable during rapid back-and-forth and mid-sentence corrections in dialogue.
- AI summary reliability (premium feature): meeting summaries are now consistently produced even when the transcript mentions a product name in quotes ("Your data. Your dashboard."), participant quotes, or code in backticks. Previously these fragments broke the response structure and the summary wasn't saved. Summaries in German, Hungarian and Korean have also become more stable.
- Translator and summary protection against content tampering attempts: added extra filtering layers for incoming data (Meet participant names, recognised speech, transcript content) so that accidental or deliberate manipulation by individual participants doesn't affect translation and summary quality for the rest of the group. Especially important for meetings with many participants or open webinars.
v1.5.3
- Multilingual meetings: MeetVoice now works on calls where participants speak different languages. The extension automatically detects each speaker's language and translates everything to yours — no need to switch settings during the meeting. A colored language badge next to the speaker's name appears in subtitles and meeting history, and PDF, SRT and TXT exports show the source language of each line
- Active-language legend in subtitles: in multilingual meeting mode, a compact strip above the subtitles shows the languages the extension is listening to. Each chip uses the same color as the badge next to the speaker's name, so at a glance you see which languages are being monitored and which one is being spoken right now
- API Keys settings made clearer: a tier overview at the top shows what each key unlocks (speech recognition, faster translation, premium voices), with colored dots that turn green as you configure them. Under each field, a live hint explains the current state — what works without the key and what you gain by adding it. Groq and OpenAI translation keys are now stored separately, so switching providers no longer mixes them up. For OpenAI users: a «Use translation key» button copies the key into the voices field with one click
- Participant names are now added to the translation glossary automatically. Translations stay consistent across the whole call — «Aleksey» no longer drifts between «Alex», «Aleksei» and «Alexey». Your manually entered glossary entries retain priority when the 500-character budget is full.
- First-install setup guide: a welcome page now opens automatically when you install MeetVoice for the first time, walking you through the two setup steps — installing the desktop application and adding your Deepgram API key. In the panel, a clear 2-step checklist replaces the old single-line warning, showing live status of both prerequisites and auto-collapsing when you're done.
v1.5.2
- Translation auto-recovery on invalid API key: if your Groq or OpenAI key stops working mid-session (expired, revoked, or mistyped), translation now seamlessly falls back to the built-in proxy for the rest of the meeting instead of failing every phrase. You'll still see a notification so you know to fix the key, but the meeting keeps going without interruption
- Quick links to obtain API keys: every key field in Settings → API Keys (Deepgram, Groq, OpenAI, OpenAI TTS) now has a «Get key →» link that opens the provider's console in a new tab — no more hunting for the right URL
v1.5.1
- Bilingual AI summary now generates correctly in all 18 languages — previously a Bulgarian, Czech, Polish or similar summary block fell back to English text with an «AI Summary» heading even though the meeting was in that language. PDF section titles («Key Decisions», «Action Items», «Topics Discussed») are now also localized for every supported language
- Refined data flow disclosure in API keys settings: keys are sent only to provider APIs (Deepgram, OpenAI, Groq), never to MeetVoice servers. The same precision now reflected in the privacy policy on the website
v1.5.0
- Keyboard shortcuts inside Meet: Alt+T toggles translation, Alt+S hides or shows subtitles, Alt+M instantly mutes the voiceover — control your session without taking your hands off the keyboard or opening the panel mid-presentation
- Meeting glossary in Settings — add coworker names, product names and technical terms (up to 500 chars), and the translator handles your specific vocabulary noticeably more accurately. Persists between sessions
- Transcript export to SRT and TXT alongside PDF — SRT for subtitle editors (Premiere, DaVinci, Subtitle Edit), TXT for quick pasting into chat or email
- AI meeting summary now works in all 18 languages — Polish, Dutch, Turkish, Japanese, Korean, Czech, Slovak, Hungarian, Romanian and Bulgarian previously failed with «Invalid language». Summaries now generate in any supported language
- 5-minute trial-ending notification with a direct link to license activation — no more sudden cutoffs mid-call
- More resilient PDF export: fixed overlapping cards in Japanese and Korean transcripts, added a safeguard against crashes on extra-long meetings (>500 phrases) by trimming to the most recent entries
v1.4.9
- Translation now auto-stops after 15 minutes of silence — protects your battery, transcription credits, and trial budget if you forget to disable it after a call
- Clearer 'Activate a license' prompt when your trial ends mid-session, with the message localized in all 18 supported languages
v1.4.8
- Added 5 new languages: Czech, Slovak, Hungarian, Romanian, Bulgarian
- Popup, subtitles, voiceover and website now available in all 18 languages
v1.4.7
- Added 5 new languages: Polish, Dutch, Turkish, Japanese, Korean
- Popup, subtitles and website now available in all 13 languages
- Auto-retry on translation rate limits — if Groq throttles, MeetVoice silently falls back to OpenAI so translation keeps flowing
- CJK subtitle rendering improvements: Japanese line-break and mid-stream sentence splitting for smoother Asian-language flow
v1.4.6
- Added 4 new languages: Spanish, Portuguese, French, Italian
- Popup, subtitles and website now available in all 8 languages
v1.4.5
- Fixed speaker detection in 1-on-1 calls — names now appear correctly even with just two participants
- Microphone off? The app no longer wastes resources processing silent audio
- Fixed a rare update notification error that could appear on app startup
- Internal cleanup to comply with Chrome Web Store review requirements
v1.4.4
- Subtitles now wrap long phrases into balanced, easy-to-read lines instead of one long string
- Subtitle font size automatically adapts to your window size — no more tiny text on small screens or oversized text on ultrawide monitors
- Fixed speaker name detection after a recent Google Meet interface update — names appear reliably again in group calls
- Fixed AI meeting summary not generating when using the desktop application
- Improved stability: resolved several timing issues that could cause brief audio glitches or subtitle flicker during long meetings
v1.4.3
- AI meeting summary now written natively in BOTH meeting languages at once — no more one-sided or machine-translated versions, each summary is crafted directly by the model in its own language
- Your PDF export now shows two clean sections back-to-back — one per language with native headings and polished formatting
- If one language hiccups, you still get the other one instantly plus a clear heads-up toast — no more all-or-nothing summaries
- Single-language meetings stay just as fast and lightweight — zero extra cost when both sides speak the same language
- Smarter error messages when summary generation fails — you'll finally know whether it's a server hiccup or an API key issue
- Deep security audit: tighter validation on all server messages, stronger API key handling, and hardened translation pipeline
- Fixed: speaker names in group calls could sometimes resolve incorrectly due to a missing internal message relay
- OpenAI TTS key test button no longer consumes API credits — uses a lightweight check instead
- Mistyped license key? You can retry in 10 minutes instead of waiting a full hour
v1.4.2
- Speaker names in group calls (3+ people) appear from the very first phrase — no more 30-second warm-up
- Wrong name guessed early? We fix it retroactively on all previous subtitle cards once the real speaker is known
- New presenter mode — much faster recognition for webinars and lectures with a single dominant speaker
- Fewer flickering labels — names only update when we're confident, no more twitchy back-and-forth on short phrases
- Fixed: speaker name corrections stopped working after waking a Meet tab from Chrome's back-forward cache
v1.4.1
- Fresh new look — wider extension panel, brand-colored layout, bigger buttons and a polished overall feel
- Turn on 'Speak my translation aloud' before you even join a meeting — no more flipping switches mid-call
- Cleaner volume controls — each slider now has a clear label so you know which audio channel you're adjusting
- Settings reorganized into four easy-to-browse sections: Voice, Subtitles, API Keys and Advanced — with a 'Setup required' hint when keys are missing
- Voice picker now uses compact play and reset icons — long voice names finally fit, and the preview button shows what it's doing
- Meeting history is easier to navigate with keyboard, and you see a loading placeholder instead of a blank screen
- Extension now starts in your browser's language automatically (Russian, Ukrainian, English or German)
- Better readability everywhere — stronger contrast on text, borders and dividers
v1.4.0
- Your voice is no longer almost inaudible while the translation plays — by default you now stay audible at 20% volume, so your peer can still hear you speak naturally behind the translation
- New slider in the panel lets you fine-tune exactly how quiet your voice gets during translation — from barely audible (10%) to no quieting at all (100%). Find it right under the Translation volume slider when outgoing translation is enabled
v1.3.9
- Pick a specific voice for each language — choose from the full neural voice catalog (English, German, Russian, Ukrainian) instead of just male or female
- Preview any voice in Settings before starting a meeting — hear exactly how it sounds in one click
- Your voice picks are remembered across sessions and survive reconnects
- Reset to default voice anytime — one button per language
v1.3.8
- Test your API keys in one click — check Deepgram, translation and voiceover keys directly from Settings
- Mute translation voice with a single tap — keep subtitles visible while silencing the voiceover
- More accurate speaker names in group calls with 3 or more participants
- Clearer error messages — when something goes wrong, you'll see exactly what to fix and where, in your language
- Improved reliability and security under the hood
v1.3.7
- Added protection against selecting the same language for both sides
- Enhanced connection security — stricter message validation
- Fixed: empty API keys in settings no longer cause connection errors
v1.3.6
- Fixed voice overlap — consecutive translations no longer play over each other
- Improved speaker name detection — works in large meetings, late joins, and 16 UI languages
- Faster voice response time
- Improved PDF export reliability
v1.3.5
- AI meeting summary — one-click summary of your entire meeting
- German language support (Deutsch) — UI, subtitles, and voice
- Improved PDF export — faster generation, cleaner layout
- Pulse indicator on transcript button while recording
- Fixed: meeting transcript now saved when session ends unexpectedly (trial expired or connection lost)
- Fixed: failed connection attempt no longer creates empty entry in meeting history
- Improved voice stability — fixed audio cutoff on consecutive speech segments
- Enhanced security and stability
v1.3.4
- Faster voice response — reduced voiceover latency
- Deepgram auto-reconnect now stops on invalid API key
- Fixed audio resources not cleaned up on mic switch in Meet
- Fixed transcript buffer overflow on very long meetings
v1.3.3
- Meeting transcript — real-time panel with speaker names and timestamps
- Export transcript to styled PDF with speaker color coding
- Meeting history — browse and manage up to 50 past meetings
- Fixed meeting time display (now uses your local time)
- All 4 languages available as user or peer language
v1.3.2
- Secure connection required for remote servers (wss://)
- Fixed meeting audio staying quiet after empty translation
- Faster audio processing with zero-copy data transfer
- Better error messages — no more confusing technical details
- Improved Telegram bot protection against spam
- Fixed date display on Download page for different timezones
- Settings labels now fully translated in RU/UK
v1.3.1
- Fixed meeting audio staying muted after translation error or reconnect
- Improved stability on unstable connections
- Enhanced desktop application license window security
- Faster translation processing via cloud proxy
v1.3.0
- Built-in translation — only 1 key needed (Deepgram)
- Translation and voice work automatically, no extra setup
- Simplified settings with collapsible advanced section
- Subtitles now appear at the top by default
v1.2.0
- Free built-in voice — no OpenAI API key needed for speech
- Only 2 API keys required now (Deepgram + Groq) instead of 3
- Male/Female voice toggle
- OpenAI TTS available as optional upgrade
- Faster voice response time
v1.1.0
- Added BYOK (Bring Your Own Keys) support
- Added UI localization (RU/UK/EN)
- Improved TranscriptBuffer with smart endpointing
v1.0.0
- Initial release: bidirectional EN/DE ↔ RU/UK translation
- Desktop app for Windows and macOS
- Chrome extension with subtitles and voiceover