Recognition language

Set the recognition language via language. Required for the whisper model.

Supported languages

• ISO 639-1 two-letter codes. See the whisper list.

Special keywords

1. detect: Detect a single language for the whole audio.
2. multi: Handle multiple languages mixed at sentence level.

language_candidates option

Used with detect or multi to narrow candidates.

• Default: ["ko", "ja", "zh", "en"]

caution

• language is required for whisper.
• Not available for sommers.
• Too many candidates may reduce accuracy.

Examples

In the POST Request body, set language (required for whisper). Optionally provide language_candidates for detect/multi.

POST example: Example

Unified example (presets): Example