Output formats
JSON, SRT, VTT, TXT.
Draft notice
Owner check needed: endpoint, parameter, limit and error names here come from the planning spec, not from the live service. Confirm them against the real API before launch.
Overview
Pick the format that skips a conversion step: JSON for code, SRT or VTT for players, TXT for reading.
- JSON, SRT, VTT or TXT.
- SRT, VTT and JSON segments have times.
- Check the first line of a saved file.
WEBVTT
00:00:00.500 --> 00:00:02.400
Okay, three tips for a small garden.
00:00:02.600 --> 00:00:05.100
Tip one: measure the sun before you buy seeds.
00:00:05.400 --> 00:00:08.200
Tip two: raised beds drain better.
00:00:08.500 --> 00:00:11.000
Tip three: water in the morning.
00:00:11.300 --> 00:00:13.600
Follow for the full guide.The four formats
| Format | Has times | Best for |
|---|---|---|
| JSON | Yes, in segments | Code, search, chunking |
| SRT | Yes | Video editors, many players |
| VTT | Yes | Web video, <track> elements |
| TXT | No | Reading, prompts to a language model |
SRT and VTT
Save the file with UTF-8 encoding. Check the first line: VTT starts with WEBVTT, SRT with the number 1. The SRT and VTT workflow does this check in code.
Not offered
Word-level timing, translation, speaker labels and Markdown are not offered unless the endpoint reference says so.
Parameters: transcript endpoint.
Ready to call it from your code?
Get API key