Free preview
Medical Dictation to Text
Drop a dictation, lecture or consultation recording and read the first 5 minutes as text, with clinical vocabulary kept intact. No account; the file is deleted once the transcript is delivered.
- Free preview: first 5 minutes
- Deleted after processing
- No signup
Drop a dictation or consultation recording here
or
Supports MP3, M4A, WAV, OGG/OPUS, AAC, FLAC and WEBM
Add SRT files (2+)
Transcript
How it works
Medical Dictation to Text in three steps
Record a dictation without identifiers
A textbook case, a lecture or your own summary. Phone close to the mouth, quiet room.
Drop it here
The browser keeps the first 5 minutes. Language is detected automatically, terms are transcribed as spoken.
Check the terminology
Copy the text or download a .txt. Everyday use, with a library and summaries, is in the app.
Dictation that comes back spelled correctly
General speech engines stumble on drug names and anatomy; a dictated "metoprolol succinate 50 milligrams" turns into something a pharmacist would query. The engine behind this page is the one that powers the SozAI app, and it handles clinical vocabulary the way it handles any other: from the audio, not from a dictionary lookup. This preview lets you check that on your own voice and your own terms before you rely on it.
Before you upload anything
This is a public preview tool. Do not upload recordings that identify a patient — names, dates of birth, record numbers, or a case described in enough detail to be recognisable. Test it with a dictated summary of a textbook case, a lecture, or your own notes with identifiers left out. Audio is processed by a US-based provider (AssemblyAI) unless the site has been switched to its EU endpoint, and the preview is not offered under a business associate agreement.
What the preview does
Your browser reads the file and keeps the first 5 minutes, cutting a longer recording locally and re-encoding that part as 16 kHz mono WAV so the rest is never uploaded. The audio goes over an encrypted connection to the transcription engine, the language is detected automatically across 99 languages, and multiple speakers — clinician and patient, lecturer and student — are labelled separately. When the text is delivered, the audio and transcript are deleted at the provider; this site stores neither.
What it is good at, and what it is not
- Dictated notes, ward-round summaries, case presentations and medical lectures transcribe close to verbatim when recorded near the microphone.
- Terminology in the language of the recording is spelled as spoken; Latin and English terms inside a Spanish or German dictation are usually kept.
- It is not a medical device, does not structure notes into SOAP or any template, and does not check dosages. Review before anything reaches a chart.
Formats and limits
- M4A, MP3, WAV, AAC, FLAC, OGG/OPUS, WEBM, up to 25 MB as-is; longer files are trimmed to 5 minutes in the browser first.
- Three previews per device per day, no account.
For daily use
The SozAI app transcribes dictation of any length, keeps every note in a library under your account, and adds summaries and translation. The first 30 minutes are free. Details of storage and encryption are on the security page.
Answers
Medical transcription — frequently asked questions
Can I upload a real patient consultation?
Not to this preview. It is a public tool without a business associate or data-processing agreement, and audio is processed by a third-party provider. Test it with de-identified dictation, and use the app for anything covered by your clinic's data rules.
How does it handle drug names and anatomy?
They are transcribed from the audio like any other words, and the engine has been trained on a great deal of clinical speech. Clearly dictated terms come back spelled correctly; mumbled or heavily accented ones may not. Always review.
Which languages are supported?
The language is detected automatically across 99 languages. Latin and English terminology inside a dictation in another language is usually preserved.
Does it structure the note into SOAP sections?
No. The preview returns the transcript as spoken. Summaries and structured notes are generated in the SozAI app; templates are a matter of your own documentation workflow.
Where is the audio processed?
By AssemblyAI over an encrypted connection, in the US region unless the site has been switched to the EU endpoint. Audio and transcript are deleted there once the text is delivered, and this site stores neither.
Is this HIPAA or GDPR compliant?
The preview makes no such claim and should not receive identifiable health data. The SozAI app's handling of recordings — storage, encryption, deletion — is described on the security page; whether that meets your obligations is a question for your organisation.
Can it transcribe a medical lecture or a conference talk?
Yes, and that is a good use of the preview: no patient data, dense terminology, one clear speaker. The first 5 minutes are free here; the full lecture is a job for the app.
Does it separate doctor and patient?
When both voices are audible it labels them as separate speakers with timestamps. Remember the rule above: only with recordings that do not identify anyone.
The whole recording, not just the first minutes
Record or import any length in the app: speaker labels, timestamps, a summary and translation into 100+ languages.
Free on iOS and Android. 30 minutes of transcription included, no card.
Free tools
More Free Tools
YouTube Transcript Generator · SozAI vs Other Tools · Alternatives to Popular Apps
