The all-in-one transcription app for Mac. Record meetings without a bot, dictate into any app, caption speech live, and turn any audio or video into text with speaker names and AI summaries. Nothing leaves your device.
Rated 4.96/5 by verified buyers · Free forever, unlimited transcription · Pro $39 once
Use code 4OH6Y0D for 15% OFF on Gumroad.
Trusted by professionals at
Netcraft
Meetings, voice, files, videos — all on your Mac.
Zoom, Teams, Meet, Slack and Discord calls recorded and transcribed automatically — no bot joins.
Hold a key and talk — polished text lands at your cursor in any Mac app.
Real-time transcription with live speaker labels and translation in 30+ languages.
Every speaker separated and labelled by name with Voice Profiles.
Drop in any audio or video, batch hundreds at once, or watch a folder for new files.
Transcribe from YouTube and 1000+ sites, then export subtitles as SRT or VTT.
The full cycle, fully automatic
Join the meeting. That’s your only job.
TranscribeX notices when a meeting starts in Teams, Zoom, Google Meet, Slack, Discord, or any macOS app, then records audio and video, matches it to your calendar event, and runs the whole pipeline on its own — with no bot in the participant list. You don’t press anything. Learn more about local meeting transcription
No clicks. No babysitting. The recap is in the inbox before you’re back at your desk.
Every recording is matched to its calendar event — named after the meeting, with date, participants and details carried into exports and the AI summary. Pick which calendars to use, or link any recording by hand.
The AI summary and full transcript are emailed the moment the meeting ends — or send any meeting manually with one click, while everyone still remembers the discussion.
Every speaker, labelled. Automatically.
TranscribeX separates and labels every voice in your recording. Teach it a voice once with Voice Profiles, and you’ll never see “Speaker 1” again — just names. Perfect for meetings, interviews, and podcasts. How speaker diarization works
See speech become text. In real time.
Watch the transcript build while meetings and recordings are still happening — and watch it sort itself out, too. Live speaker diarization labels each voice as it speaks, so the transcript is already attributed by the time the call ends. New streaming models cover 30+ languages in real time, with punctuation and capitalization as they go, and headphones are optional — TranscribeX keeps its own audio out of the transcript.
An hour of audio. The gist in seconds.
Turn any meeting, lecture or interview into a report: key points, decisions, and action items, with your own notes woven in. Chat with your transcripts to ask “what did Priya commit to?”, and reuse predefined or custom prompts for the jobs you do every day.
Hold a key. Speak. Watch it typed.
Dictate into any Mac app — hold a key of your choice, talk, release, and the text lands right at your cursor. AI polish strips the “um”s, fixes the “Tuesday… wait, no, Wednesday” moments, and formats numbers and lists — while the speech itself never leaves your Mac. fn is all it takes.
Bring your own AI. Or run one built in.
Every AI feature in TranscribeX — summaries, chat, dictation polish, translation — runs on whichever brain you choose.
Plug in any major AI service, or point TranscribeX at any OpenAI-compatible endpoint. Your keys, your models, your choice — switch providers any time.
One click downloads a local model that runs right inside the app. No API key, no account, no subscription, no per-token bill — and your transcripts never leave your Mac.
Pick the engine that fits. Or try them all.
Every model downloads once and runs entirely on your Mac — switch engines per job, per language, or per mood. Compare the models
The accuracy benchmark. Every size from Tiny to Large V3 Turbo, transcribing 100+ languages with best-in-class precision.
The speed engine. Up to 20× faster than Whisper with balanced accuracy, plus low-latency streaming variants.
The live engine. Streaming recognition in 30+ languages with punctuation and capitalization as you speak — the model behind live captions and live diarization.
Modern multilingual recognition with a forced-aligner companion model for precise word-level timestamps.
The speech engine built into macOS — nothing extra to download, instant startup, and 30 locale variants.
Fine-tuned Whisper — optimized for your language
Community fine-tunes of Whisper Large V3, each trained specifically for one language — noticeably better than the general model on accents, names, and local phrasing.
The list keeps growing — don’t see your language? Request it and we’ll add it.
Drop a file. Any file. Or a whole folder.
Drag audio or video straight into TranscribeX — or point it at a folder and it transcribes new files as they arrive. Queue hundreds at once with batch transcription, then play everything back with the built-in player.
Audio
Video
Export to
Everything else
From verified customer reviews on Gumroad
“TranscribeX has been great! I am on 20+ calls weekly and this tool has become invaluable for call transcripts, and without any subscription or sending my data off to the cloud! I've emailed support on a few occasions and Ethan has been extremely responsive, top notch support!”
“TranscribeX really surprised me—fast, accurate, and private. Love that it runs local AI with no extra costs for transcription.”
“The app is intuitive and well done. The developer keeps actively developing the app, and I enjoy the wide support of various transcription models. Plus being able to download videos, create subtitles and translate them within the same app, with just a few clicks — this is a great piece of software!”
“I've bought it and was very happy with the performance. Now I suddenly got the ability to do dictation which makes me very happy too. Keep improving it! Thanks Ethan”
“The developer is incredibly responsive, fixed my reported issue in record time! The app now works beautifully, is lightning fast, and the experience overall is seamless. Truly impressive support and a fantastic product.”
“great little app—get off the cloud ;)”
Free forever — unlimited transcription. Upgrade when you need more power.
No subscriptions. No per-minute fees. Otter Pro costs $99.96/year — TranscribeX Pro is $39 once, yours for life.
Otter Pro price billed annually, as listed on otter.ai (checked October 2026).
Use code 4OH6Y0D at checkout
| Feature | Free | Pro |
|---|---|---|
| Transcription | ||
| Unlimited transcription & export | ||
| Import audio & video files | ||
| Record with Mac microphone | ||
| 100+ languages | ||
| YouTube & 1000+ sites download | ||
| Batch transcription & export | — | |
| Folder auto-detect and transcription | — | |
| Bulk YouTube download (HD) | — | |
| Word-level timestamps | — | |
| Transcript translation (Apple Translate, DeepL, AI) | — | |
| Models | ||
| Whisper Tiny & Apple On-Device | ||
| All Whisper & Qwen3 ASR models | — | |
| NVIDIA Parakeet (20× faster) | — | |
| NVIDIA Nemotron live models (30+ languages) | — | |
| Fine-tuned Whisper models | — | |
| Online models: ElevenLabs Scribe & Mistral Voxtral | — | |
| Meetings | ||
| Detect & record Zoom, Teams & Meet calls (no bot) | ||
| Record audio & video from any macOS app | — | |
| Fully automatic: transcribe, diarize, summarize & export | — | |
| Email summaries & transcripts (auto or one click) | — | |
| Calendar integration & event matching | — | |
| Speakers & live | ||
| Speaker diarization & Voice Profiles | — | |
| Live transcription with live speaker labels | — | |
| Real-time translation | — | |
| Dictation | ||
| Dictation in any Mac app | ||
| AI polish for dictation | — | |
| AI | ||
| AI summaries (local LLM, ChatGPT, Gemini, Ollama…) | — | |
| Chat with your transcripts | — | |
| Predefined AI prompts | — | |
| Editing & export | ||
| Edit & merge transcript segments | ||
| Search & replace, dictionary, tags | ||
| Built-in media playback | ||
| Export to TXT | ||
| Export to PDF, SRT, VTT, CSV & Markdown | — | |
| CLI & MCP automation | ||
| Support | ||
| Priority support + 7-day refund | — | |
Unlimited free transcription. Pro is $39 one-time — no subscription.