Real-time Transcription
Learn about Tinrec’s real-time transcription: transcribe while recording, auto-detect language, and generate transcripts and meeting notes live.
What is real-time transcription
Real-time Transcription converts speech to text while recording is in progress and shows the transcript live. Unlike “record first, transcribe later,” real-time transcription lets you see text as soon as you speak—no post-processing wait.
Tinrec is an AI meeting assistant with the tagline “One-tap meeting notes · AI meeting assistant.” Real-time transcription is one of Tinrec’s three transcription entry points; the other two are Audio to text (upload existing files) and Web video to text (paste a video link). They serve different purposes—see “Feature boundaries” below.
How real-time transcription works
Tinrec’s real-time transcription follows a “transcribe while recording” model:
- Transcribe while recording: After you start recording, the system continuously converts speech to text and shows it on screen; the transcript appears in sync with speech.
- Start speaking to record—no manual language switch: You don’t need to pick a language before recording. The system defaults to auto-detect and supports Chinese, English, Japanese, Korean, German, Cantonese, Taiwanese, and more, with ongoing expansion.
- Desktop captures meeting system audio: The Mac (macOS 12 or later) and Windows (Windows 10 or later) desktop apps can record system audio from Zoom, Meet, Teams, Webex, and similar apps by capturing device system audio—not by joining the meeting as a bot.
- Auto output after stop: When you stop recording, the system automatically generates meeting notes and to-dos—no manual cleanup.
Note: Offline transcription support is as stated on the official website.
Use cases and related capabilities
Use cases
Real-time transcription fits situations that are happening now and need on-the-spot notes, such as:
- Meeting recording (online or in person)
- Class and lecture recording
- Interviews and focus groups
- Call recording
Related capabilities
Beyond the transcript itself, real-time transcription can work with these features (availability varies by platform):
- Live translation
- Live captions
- In-app audio capture on mobile
- Bilingual side-by-side display
- Adjustable transcript font size
Tip: Some of these capabilities vary by platform. Available options depend on the App interface on your device.
Feature boundaries: which transcription to choose
Tinrec offers three transcription entry points for different situations:
| Entry point | Best for | In one sentence |
|---|---|---|
| Real-time transcription | What’s happening now | Transcribe while recording and see the transcript live |
| Audio to text | Files you already have | Upload an existing recording and transcribe afterward |
| Web video to text | Someone else’s video link | Paste a video link and transcribe online video content |
In short: real-time transcription handles sound in progress, audio to text handles existing files, and web video to text handles other people’s video links.
Supported platforms and allowance overview
Supported platforms
- iOS, Android, Mac desktop (macOS 12 or later), Windows desktop (Windows 10 or later), and Web are all available.
- Both Mac and Windows desktop can record system audio from Zoom, Meet, Teams, Webex, and similar apps.
Allowance overview
- Free: 60 minutes per month (no credit card required); max 30 minutes per recording (web Free).
- Pro: Longer single sessions, up to 5 hours; exact web per-session minutes as stated on the official website.
- Allowance details: Basic Pro about 600 minutes and up; Pro+ 1200 minutes/month (details as stated on the official website).
Accuracy
Real-time transcription accuracy is about 98–99%. Actual accuracy depends on background noise, pronunciation clarity, language, and other factors.
Related actions
Links to detailed how-to guides: