How to turn an MP4 into a transcript
- Choose your MP4. Drag the file onto the box above or tap to pick it. Free files can be up to 120 minutes long.
- Pick the language and speaker labels. Choose the spoken language (30+), or leave it on auto. Turn on speaker labels for interviews and meetings.
- Wait about a minute. A one-hour file is usually done in under a minute. Keep the tab open or come back from My files.
- Download the transcript. Copy the text or download TXT, DOCX, SRT or VTT. Rename speakers first and the names appear in every export.
Limits, formats and prices
| Free | Pro | Minute packs | |
|---|---|---|---|
| Minutes | 120 a month | 1,500 a month | 120 min $6 · 400 min $15 · 1000 min $29 |
| Longest file | 120 min | 3 h | 3 h |
| Price | $0, no card | $12/month or $96/year | One-off, valid 12 months |
| Speaker labels, 30+ languages, all four exports | Yes | Yes | Yes |
Also accepted: MP3, M4A, WAV, AAC, OGG, OPUS, FLAC, WebM, MP4, MOV, MKV and M4V. Minutes are counted on the audio length; if a file fails, the minutes come back.
Which export to choose
| You want to… | Export |
|---|---|
| Read, quote, or paste into ChatGPT or Notion | TXT |
| Edit and share a document with speaker names | DOCX |
| Burn or attach subtitles in Premiere, Resolve or CapCut | SRT |
| Captions for a web player | VTT |
Before and after: what the transcript looks like
An MP4 has no text inside it — just sound. The transcript is built from the audio. For a two-person interview with speaker labels on, the result reads like this (format shown with placeholder content):
| Before | After (TXT with timestamps) |
|---|---|
| interview.mp4 — 42 min, two voices, no captions | [0:00] Speaker 1: Thanks for coming in. Let's start with how the project began. |
| [0:06] Speaker 2: It started as a side experiment in 2023… | |
| Rename Speaker 1 → Host, Speaker 2 → Guest | [0:00] Host: Thanks for coming in… · [0:06] Guest: It started as… |
The SRT export of the same file keeps one cue per spoken line, with the speaker name at the start, ready to import as subtitles.
Accuracy tips
- Set the spoken language instead of auto when the first minute is music or silence.
- One microphone per speaker helps speaker labels more than anything else.
- Background music lowers accuracy; a clean voice track (or an export without music) is best.
- For a YouTube video that already has captions, paste the link on the home page instead — it is instant and costs no minutes.
Privacy
Your MP4 is uploaded over HTTPS, transcribed and deleted when processing finishes — never kept longer than 7 days, never used to train models. The transcript stays in My files until you delete it.
What it can and can't do
Can
- Transcribe MP4 and MP3, M4A, WAV, AAC, OGG, OPUS, FLAC, WebM, MOV, MKV and M4V files in 30+ languages.
- Label and rename speakers.
- Export subtitles (SRT, VTT) with timing from the audio.
- Summarize the finished transcript into key points and chapters.
Can't
- Take files longer than 120 min on the free plan (Pro: up to 3 h).
- Read text that only appears on screen in the video.
- Guarantee perfect names and jargon — check proper nouns before you publish.
Frequently asked questions
Can I convert an MP4 to a transcript for free?
Yes — 120 minutes every month with no card, files up to 120 minutes.
How long does it take?
Usually under a minute for a one-hour file. Longer files take a little more.
Can I get SRT subtitles from an MP4?
Yes. Download SRT (or VTT) from the finished transcript; the timing comes from the audio.
Does it tell speakers apart?
Yes. Turn on speaker labels before uploading; rename Speaker 1, Speaker 2… and the names appear in every export.
Is my video kept?
No. Files are deleted when processing finishes (7 days at most) and are never used for training.
What if the MP4 is longer than 2 hours?
Free files can be up to 2 hours; a minute pack or Pro takes files up to 3 hours. Or split the file and upload the parts.