Accepts a complete transcript-package response with media duration up to 1800 seconds, including words/SRT/VTT/metadata, or legacy speaker segments. Uses every cue; adjacent same-speaker cues are joined with one space, retaining original cue index ranges. Anonymous labels stay anonymous. Up to 20000 input cues/segments and 120000 normalized text characters; 16 MiB JSON transport bound.
| Network | Scheme | Amount | Pay To |
|---|---|---|---|
| Base | exact | $0.009000 USDC | 0x2880...4E5E |