Transcribe a Recording into a PDF, on Your Device
Transcribe audio, video or a mic recording on your device with Whisper, then save it as PDF, TXT or SRT.
Free, with no sign-up and no watermark. The file is processed in your browser and never uploaded.
Transcribe an audio or video file — or a recording from your microphone — into a PDF transcript. Speech recognition runs on your device with OpenAI's Whisper model: choose Fast (about 64 MB to download the first time) or Better (about 105 MB). Pick the spoken language, review and edit the transcript, then download it as PDF, .txt or .srt subtitles.
Step by step
- Add audio: Upload an MP3, WAV, M4A, OGG, WEBM or MP4 file, or click Record with microphone.
- Choose accuracy and language: Set Accuracy and Spoken language.
- Transcribe: Click Transcribe to PDF. Long recordings take a while on slower devices.
- Edit: Correct names and terms under Transcript — edit before exporting. Tick Include timestamps in the PDF if you want them.
- Download: Click Download PDF, or save .txt or .srt subtitles.
Handy for
Interviews
Get a transcript to quote from.
Lectures
Turn a recorded class into study notes.
Meetings
Keep a written record of a call.
Videos
Create subtitles from a video's soundtrack.
Good to know
Clear audio with one speaker at a time transcribes best.
The PDF supports Latin, Cyrillic, Greek and Vietnamese scripts. For Indic, Arabic or CJK transcripts, use the .txt download.
Speakers aren't labelled — add names while editing.
Questions
Are my files uploaded anywhere?
No. The file is read and processed by JavaScript running in your browser tab. Nothing is sent to a server, and the result is created on your device.
Why does it download something the first time?
On-device AI needs a model file. It is downloaded once from a public model host, cached by your browser, and then runs locally — your document itself never leaves your device.
How long can the recording be?
There's no fixed limit, but transcription runs on your device and long files can take many minutes.
Further reading
- Transcribe Audio to Text Privately, on Your Own Device — Speech to Text runs OpenAI's Whisper model in your browser: pick the language, transcribe, then export text, SRT/VTT subtitles or a PDF.
More Make a PDF tools
- Typeset CV Maker — Fill in a form, pick a Classic, Modern or Sidebar layout, preview it live and download a typeset resume PDF.
- Convert Anything to PDF — Turn images, HEIC photos, Word, Excel, PowerPoint, HTML, text and EPUB files into PDF — mix them and merge into one.
- JPG & Photos to PDF — Place JPG, PNG, WEBP or other images one per page on A4, Letter or image-sized pages, in the order you choose.
- Word (DOCX) to PDF — Convert .docx files to PDF with text, lists, tables, images and page breaks, keeping the text selectable.
- Excel Sheet to PDF — Turn .xlsx, .xls, .ods or .csv sheets into PDF tables, keeping merged cells, fills and column widths.
- PowerPoint Slides to PDF — Draw each .pptx slide as a PDF page at its own size, with selectable text and a list of anything skipped.