Turn recordings into text your users can use.
Send an audio file from your app; get back a transcript with language and timestamps. Identity, subscription status and usage limits are checked on the way — you only handle the result.
Okay, quick note for tomorrow: move the stand-up to ten,and remind me to send the invoice before lunch.
Voice notes
Record, upload, read. Searchable text from the first memo — the pattern behind Owlly.
Meetings and lectures
Longer recordings with timestamps, so the app can jump to the moment a sentence was said.
Voice input
Short clips turned into text for a search box, a chat message or a form field.
One upload. One result.
- 1Upload the audio file with the user token.Multipart upload from the device. No provider key in the app.
- 2Backapp checks the user and runs the job.Identity, subscription status and usage limit are verified first; declined requests never reach the provider.
- 3Read the transcript from the job.Poll the job until
completed, then show the text — and the timestamps if you want them.
// CONTENT PENDING: replace with the verified example
const form = new FormData();
form.append('file', { uri, name: 'recording.m4a', type: 'audio/m4a' });
const res = await fetch(`${BACKAPP_API_URL}/ai/transcriptions`, {
method: 'POST',
headers: { Authorization: `Bearer ${userToken}` },
body: form,
});
const job = await res.json();
// poll until completed
const result = await waitForJob(job.id);
setTranscript(result.text);{
"id": "job_01HZX…",
"status": "completed",
"workflow": "transcription",
"result": {
"text": "Okay, quick note for tomorrow: move the
stand-up to ten, and remind me to send
the invoice before lunch.",
"language": "en",
"duration": 14.2,
"segments": [
{ "start": 0.0, "end": 6.8, "text": "Okay, quick note for tomorrow: move the stand-up to ten," },
{ "start": 6.8, "end": 14.2, "text": "and remind me to send the invoice before lunch." }
]
}
}What comes back.
A job with the transcript and enough structure to build on — not just a blob of text.
text- The full transcript as one string.
language- Detected language code, e.g. "en".
duration- Audio length in seconds.
segments[]- Timed chunks with start, end and text.
status- queued, processing, completed or failed.
Transcript with language and timestamps
Plain text plus timed segments, with the detected language. Common mobile recording formats are accepted — the exact list and maximum length are in the docs.
Speaker labels and summaries
Who-said-what and an AI summary of the transcript are being scoped and will be listed here when confirmed.
The transcription behind our own note-taking app.
This workflow is the one Owlly runs on. Same API, same limits, same subscription check.
Questions about transcription?
Which audio formats are accepted?
Common mobile recording formats (such as m4a, mp3 and wav). The exact list and maximum file length are in the docs.
Which languages are supported?
The language is detected automatically and returned with the transcript. The supported-language list is in the docs.
Is the transcript returned in real time?
No. Transcription runs as a job: upload the file, poll until completed, read the result. Streaming is not part of the current scope.
Does a declined request count against the limit?
No. Identity, subscription and limit checks happen before the job is created. Only accepted jobs count.
Who pays for the AI provider?
Either you, with your own provider key in your Backapp project, or Backapp, billed on usage. Both options are described on the main page.
Add transcription to your app.
Identity, subscriptions and AI workflows from one backend.