GET

/v1/schemas/fal/audio/fal-ai%2Fminimax%2Fpreview%2Fspeech-2.5-hd

kind inputactivity audiohash 33fa2242extracted Jul 6, 2026version currentsource upstream spec ↗

Request body

6 properties · 1 required · 3 $defs
propertytypedescriptionconstraints
text*stringText to convert to speech (max 5000 characters, minimum 1 non-whitespace character)length ≥ 1 · length ≤ 5000
output_formatenumFormat of the output content (non-streaming only)url | hex · default: "hex"
language_boostenumEnhance recognition of specified languages and dialectsPersian | Filipino | Tamil | Chinese | Chinese,Yue | English · +35 more · nullable
pronunciation_dictPronunciationDictCustom pronunciation dictionary for text replacementnullable
voice_settingVoiceSettingVoice configuration settings
audio_settingAudioSettingAudio configuration settings

Validate a payload

before you spend tokens
shell
$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"fal","endpointId":"fal-ai/minimax/preview/speech-2.5-hd","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}