POST /api/v1/client/tasks with the same authentication as other models. The model is always V6 Mini. Do not send model inside input.GET /api/v1/client/tasks/{taskId}, same as Suno music generation.customMode. Use it to switch Simple / Advanced. Fields that are not allowed in the current mode return a parameter error if sent. Hidden form fields must not be submitted.customMode: false):customMode=false, prompt (scene description, β€3000 characters)promptscript, tone, variety, vocalGender, or backgroundMusiccustomMode: true):customMode=true, script (spoken text, β€5000 characters)tone (β€1000), variety (0β4, default 1), vocalGender (f / m, omit for auto), backgroundMusic (default true)promptvariety is the same as music generation: display offβmax, values 0β4. Sending 2 or "2" both work. 0 off, 1 normal, 2 high, 3 extra, 4 max.waiting / queuing / generating β possibly suno_text_success β success / failsunoData usually has 2 variations. Each item includes:id (audioId)audio_url (final downloadable audio)stream_audio_url (streaming URL)image_url, durationprompt (spoken script), tags (tone; upstream may rewrite to English), titlemodel_name=chirp-goose, is_speech=truecurl --location 'https://api.emix.ai/api/v1/client/tasks' \
--header 'Content-Type: application/json' \
--data '{
"model": "suno/speech",
"callBackUrl": "https://xxx/cb",
"input": {
"customMode": false,
"prompt": "A friendly out everyday science."
}
}'{
"code": 422,
"msg": "success",
"data": {
"taskId": "task38695****935u9043u"
}
}