Built for how reporters work
- Drop the file, not a settings page. The spoken language is detected automatically.
- Speaker labels on request. Clean turns for panels and two-way interviews.
- Timestamps when you quote. Hidden by default; shown when you ask.
- Two hours, no splitting. Long tape is processed in sections and joined back together.
Everything you get
| Output | Details |
|---|---|
| Text | Spoken language or any of 30 output languages |
| Article | Headline, lead, sub-heads, quotes and context — on a separate click |
| Exports | TXT, Markdown, DOCX, PDF, SRT with timecodes |
| Records | Text and articles saved and searchable; audio is deleted after processing |
Accuracy, honestly
Clear audio gives strong results. Noise, crosstalk, heavy accents and unfamiliar names are where any AI slips. Use timestamps to verify quotes against the recording — our 7-check list takes a few minutes and keeps you safe.
Frequently asked questions
How long can a recording be?
Up to 2 hours per recording and 1 GB per file. Long recordings are processed in sections and joined back together, so the upload doesn't fail halfway. Large videos are compressed on your own device first, so only a small audio file is uploaded.
What do I get after converting a recording?
Clean text in your chosen language with optional speaker labels and timestamps, and an article draft written from that text: headline, lead paragraph, sub-headings, quotes and context. Export as TXT, Markdown, DOCX, PDF or SRT.
Is the article ready to publish as it is?
Treat it as a strong first draft. AI can mishear names, numbers and heavy accents, so check every quote, figure and proper noun against the recording before you publish. You are the editor; AI Journalist removes the typing and translating, not the responsibility.
What happens to my audio file?
Audio is deleted from the server as soon as the text is produced. Your records keep the text and the article — not the recording.