How to transcribe a file with Kotosay
What happens when you transcribe a recording
To transcribe a recording is to write down what was said. On Kotosay it is AI transcription. Your recording goes to secure servers, the speech is turned into text, and the transcript opens in your browser with a timestamp on each line.
Everything runs on the website, in any browser, with nothing to install. Audio and video both work. With a video, your browser pulls out the sound and compresses it first, so even a large screen recording usually sends only a small audio file.
90+ languages, detected for you
Leave the language on automatic and Kotosay detects it before it transcribes the file. If you already know it, or the file opens with a minute of music, pick the language from the list before you upload.
Kotosay works in 90+ languages, from English and Spanish to Arabic and Japanese.
What you can do before you pay
Your first file needs no account and shows its first 5 minutes in full, enough to judge your own recording before you share an email address.
- A free account transcribes 3 files a day.
- Each free file shows its first 5 minutes. A longer file still uploads, and the rest stays locked until you open it with Single file or Unlimited.
- The editor is free on every plan, even on your first file without an account.
- Exports and translation come with Single file and Unlimited.
Unlimited pricing and what it adds
Unlimited costs $9.99 a month, or $119 if you pay once a year. It suits people who transcribe every week or work with long files, such as full lectures, webinars or podcast episodes.
- Files up to 10 hours long
- 3 files processing at the same time
- Use on up to 3 devices
- TXT, DOCX, PDF, SRT and VTT downloads
- Translation that keeps every timestamp
TipYou can cancel in one click from your account. Your renewal date is in your account at any time, so you can decide in good time.
AI or human transcription
AI transcription costs far less than a typist and runs at any hour, weekends included. It fits interviews, lectures, meetings and voice memos, and you can fix any word yourself in the editor.
A human transcription service has a person type out the recording. People choose it for court records and similar work where every line has to be signed off. With AI text, check names and numbers against your own recording before you publish.
What decides the quality of a transcript
The recording matters most. One speaker close to the microphone in a quiet room is transcribed cleanly. Music under the voice, people talking over each other and a phone at the far end of a table all cost words.
TipUse the free first 5 minutes as a test. If they read well, the rest of the same recording usually will too.
Your file and your payment
We delete your audio file automatically 24 hours after transcribing it. Your transcript stays until you remove it. Delete a single file, delete the whole account, or ask support to delete your data for you.
Payments for Unlimited go through a secure checkout. If a file fails or something looks wrong, write to support. We answer within 24 hours.
