AI translation refinement

A second model pass rewrites the translation with surrounding context.

The first translation optimises for latency. With aiEnhance a second pass revisits it using the surrounding segments and your glossary.

Request

json

{  "translate":  { "languages": ["en", "zh"] },  "aiEnhance": true}

Fields

Field	Type	Description
`aiEnhance`	`boolean?`	Second-pass model refinement of each translation. Default `false`. Requires `translate`.

What you receive

json

{  "type": "translation.final",  "segmentId": "gw_8f3a_2",  "text": "零点附近的量化台阶大约是 8。",  "enhanced": true}

The refined version is pushed again under the same segmentId, with enhanced: true. Replace the earlier text keyed on segmentId. The first translation is not delayed by this.