Fact · Languages and accents
OpenAI's speech-to-text guide reserves its specialized models for cases such as subtitle formats and translation into English.
Checked against its source on October 17, 2026
In the source's own words
“Use a specialized model only if you need speaker labels, word timestamps, subtitle formats, or translation into English.”
Speech to text (OpenAI API guide), OpenAI, OpenAI, 2026, Opening section on model choice.
Open the original- What it measures
- The guide recommends its newer transcription model for ordinary recordings and names whisper-1 for the subtitle and translation cases.
- What it does not show
- The guide's model advice changes over time. Check the current model table before you choose a model for a subtitle project.