Upload a ZIP of recordings and get every file transcribed
Upload one .zip with up to 100 audio or video files inside, folders and all, and each recording is transcribed in full with its own timestamps. You get back a ZIP with a TXT and an SRT for every recording, and one merged document. A batch of up to 15 minutes is free in full; otherwise $5 covers up to 6 hours of audio in total.
- Folders inside the ZIP are fine; files are ordered by their full path
- Files that are not audio or video, and macOS "__MACOSX" folders, are skipped
- Up to 100 recordings and 4 GB per upload
- Duplicates are recognized by content and counted once
- A results ZIP with numbered TXT and SRT files, one pair per recording
Why upload a ZIP
Recordings usually already live in a folder: an export from a voice recorder, a shared drive of interviews, a dictation app backup, call recordings copied off a phone. Compressing the folder and uploading one file is quicker than picking files one by one, and it keeps the folder structure, which decides the order of the results.
What happens to the contents
- Every audio and video file is found, including those in subfolders
- PDFs, images, notes and anything else that is not audio or video are ignored
- The "__MACOSX" folder a Mac adds when it compresses a folder is left out
- Two copies of the same recording, even under different names, are transcribed and counted once
- Files are sorted by folder path and name, so "Day 1/01.mp3" comes before "Day 2/01.mp3"
What comes back
You get a ZIP of results: for every recording a TXT transcript and an SRT subtitle file, numbered in order, such as "01 name.txt" and "01 name.srt". The merged transcript is in there too. You can also download one combined Word, PDF or TXT file that starts with a list of all recordings and their durations, then gives each one under its own heading.
Timestamps such as [12:40] count from the start of each recording, not from the start of the batch. Speech is written word for word in the language spoken, with [inaudible] where a word cannot be made out, and speaker changes are marked "Speaker 1:", "Speaker 2:" when they are clear.
Size and price
One upload can be up to 4 GB. You pay for the total length of audio: one payment of $5 covers up to 6 hours across all the files, and Unlimited, $19 a week, covers up to 12 hours in total per job. If the folder holds more than that, split it into two ZIPs and run two jobs. Before paying, the first part of the first recording is shown free on screen.
How the tool works, what it costs and how your files are handled is on the Bulk Audio Transcription page; prices for every tool are on the pricing page.
Last updated
Questions and answers
Can ChatGPT open a ZIP of audio files and transcribe them?
An assistant may be able to unpack a ZIP, but turning many recordings into full transcripts in one chat tends to hit audio length limits and reply length, so results come back partial. This tool processes every recording and returns a transcript for each.
What if my ZIP has other files in it?
They are skipped. Only audio and video files are transcribed, and the "__MACOSX" folder is ignored.
Does the folder structure matter?
It sets the order. Files are sorted by their full path, folder names included, and numbered in that order in the results.
How much does a ZIP of recordings cost?
$5 for the whole batch, up to 6 hours of audio in total. A batch of up to 15 minutes in total is free in full.
Can I paste a link to a ZIP instead?
No. Bulk jobs take uploads only, so download the ZIP from cloud storage first and upload it here.