A recording is unreadable until it is a transcript, and most transcription services want the recording uploaded to get there. This does the recognition on your device instead. The audio is decoded and resampled, a speech model runs in a background worker, and the fragments it produces — one every few seconds — are joined into paragraphs at sentence boundaries rather than left as a wall of one-line entries. Timestamps can head each paragraph, and the result exports as a PDF, as plain text, or as SRT and VTT subtitle files that carry the timing with them.
Features
Recognition runs on your device — the recording is never uploaded
Three model sizes, from a fast English-only one to a slower multilingual one
Fragments merged into paragraphs at sentence ends, not left one line each
Timestamps at the head of each paragraph, switching to hours when needed
Export as PDF, plain text, SRT or VTT
Page size, margins and typeface under your control
Works on the audio track of a video file too
How to use the Audio to PDF Transcript
Drop in an audio or video file
Choose a model and, if it is multilingual, the spoken language
Wait while it transcribes — roughly real time on most machines
Add a title, then export the PDF or the subtitle files
Example
Input
board-call.m4a — 34 minutes, base multilingual model
Output
412 paragraphs, 6,140 words, exported as board-call.pdf and board-call.srt
Timestamps switch to hours automatically once a recording passes sixty minutes.
Common errors & troubleshooting
The browser could not decode the audio. — Try MP3, WAV, M4A or OGG. Some containers and codecs are not decodable in a browser at all; convert the file first.
Names and technical terms are wrong. — That is where recognition is weakest, and no model fixes it entirely. A larger model helps; reading the transcript against the recording is still necessary.
It ran out of memory on a long recording. — Split the file into shorter pieces, or choose a smaller model. A two-hour recording at the small model size is beyond most browsers.
The language came out wrong. — Set it explicitly instead of leaving detection on. The English-only model always assumes English, whatever it hears.
Frequently asked questions
How accurate is the transcript?
Good on clear single-speaker audio, noticeably worse with crosstalk, strong accents, background noise and unfamiliar names. Treat it as a draft to correct rather than a record to rely on.
Does the recording get uploaded?
No. It is decoded and recognised in a worker on your device. The model weights are downloaded once from a public model host and then cached by your browser.
How long does transcribing take?
Roughly real time on a typical laptop with the small model — a thirty-minute recording takes about thirty minutes. Larger models are slower and more accurate.
Which model should I choose?
The English-only one for clear English, since it is smallest and most accurate on that. The multilingual base model for other languages or heavy accents, and the small model when accuracy matters more than time.
Can I get subtitles rather than a document?
Yes — SRT and VTT are both exported, with the timing from the recognition, so they can go straight onto a video.
Related tools
PDF Read Aloud — Listen to a PDF read out loud, passage by passage.
Text to PDF — Turn typed or pasted text into a paginated, downloadable PDF.
PDF Summarizer — Summarise a long PDF section by section, on your device.
PDF to Text — Extract selectable text from a PDF as plain text or Markdown.
Markdown to PDF — Render Markdown — headings, lists, tables and code — as a PDF.
PDF OCR — Recognise the text in a scanned PDF and make it searchable.
Chat with PDF — Ask questions about a document with a model that runs on your own device.
More PDF tools
Image to PDF — Combine images into a single PDF (auto / portrait / landscape).
PDF to Images — Render PDF pages to PNG or JPG images you can download.
Merge PDFs — Combine multiple PDFs into one, with reordering.
Split PDF — Extract pages or ranges (e.g. 1-3,5,7-9) into a new PDF.
Rotate PDF — Rotate all pages or a selection by 90/180/270°.
Organize PDF — Reorder, rotate and delete pages from thumbnails, then export.
Add Page Numbers — Stamp page numbers onto a PDF (position, format, start).
Watermark PDF — Add a diagonal text watermark to every page.
Crop PDF — Trim margins from every page via the crop box.
PDF to Text — Extract selectable text from a PDF as plain text or Markdown.
Compare PDFs — Diff the extracted text of two PDFs line-by-line.
Encrypt PDF — Password-protect a PDF with AES-256 or RC4 encryption.
Decrypt PDF — Remove password protection from an encrypted PDF.
Compress PDF — Shrink a PDF by re-rendering each page to a JPEG at a chosen quality and resolution.
PDF Metadata Editor — View, edit or strip a PDF’s title, author, subject, keywords and producer.
Resize PDF — Rescale every PDF page to A4, Letter or another standard size, content centred.
Flatten PDF — Lock filled form fields into the page so a PDF looks the same everywhere.
PDF N-up — Print 2, 4, 9 or 16 pages per sheet, with a booklet option for folding.
Interleave PDF Pages — Merge odd and even scan passes back into one document, reversing if needed.
PDF to Single Long Page — Stack every page onto one continuous sheet you scroll instead of paging.
PDF Form Data Extractor — Pull the answers out of a filled PDF form as JSON or CSV, field by field.
Blank Page Remover — Find and delete blank pages from a scan, with a tolerance you control.
Text to PDF — Turn typed or pasted text into a paginated, downloadable PDF.
Markdown to PDF — Render Markdown — headings, lists, tables and code — as a PDF.