Transcript
The transcript will appear here.
Record, transcribe, and understand every voice.
Record or upload, then turn speech into a transcript you can work with.
- Audio is processed in memory and not stored on our servers
- 30 free transcription minutes every day, up to 180 per month
- English, Chinese and Spanish
Ready
30 free minutes left today
00:00
Ready
Tap Record and your browser will ask for microphone access. You can also upload a file.
Microphone access is blocked
Your browser is blocking the microphone for this site. Follow the steps below to allow it again, or upload an audio file instead.
Playback
00:0000:00
Trim
Segments
Advanced
BYOK mode sends your audio through our proxy to Deepgram using your own API key. The key is never stored on the server. Speaker labels depend on audio quality and may need manual correction.
Audio stays on this device until you press Transcribe.
Workspace
How it works
- 01
Record or upload
Use your microphone or drop in an existing audio or video file.
- 02
AI transcription
Speech is transcribed with timestamps and optional translation.
- 03
Turn it into output
Get a summary, action items, show notes, quotes and subtitle files.
Built for
Interviews
Journalists, researchers and podcast hosts who need clean quotes and timestamps.
→Podcasts
Turn an episode into show notes, chapters, highlights and subtitles.
→Meetings and voice memos
Get the decisions, owners and next steps without listening back.
→Study and research
Lecture and interview recordings become searchable, exportable notes.
→Frequently asked questions
Is the transcription really free?
Yes. Every device gets 30 free minutes per day and up to 180 minutes per month, powered by Cloudflare Workers AI. When the daily or shared capacity is used, the tool asks you to come back later or upgrade.
Do you store my recordings?
No. Audio is processed in memory for the transcription request and is not written to object storage. Transcripts stay in your browser unless you export or save them.
How accurate is it?
Whisper Large v3 Turbo is strong for clear speech and common languages. Background noise, heavy accents, overlapping speakers and unusual names still need review.
Does it support Chinese and Spanish?
Yes. The interface and content are available in English, Simplified Chinese and Spanish, and you can transcribe more than 90 languages.
Can I get speaker labels?
The free tier produces timestamps and segments. You can add speaker labels today by using your own Deepgram API key in Advanced options.
Recommended tools
If Loqena is not the right fit, these are established alternatives. Some links are affiliate links.