Fact · Transcription
OpenAI advises against splitting a long recording in the middle of a sentence, because the cut can remove context and reduce accuracy.
Checked against its source on October 17, 2026
In the source's own words
“Avoid splitting in the middle of a sentence, which can remove context and reduce accuracy.”
Speech to text (OpenAI API guide), OpenAI, OpenAI, 2026, Longer inputs section.
Open the original- What it measures
- OpenAI's guide gives this advice for files larger than its 25 MB upload limit, where the recording must be compressed or split into chunks before it is sent.
- What it does not show
- The guide states the advice without measuring the accuracy loss. The size of the loss depends on the model and the audio.