Qwen 3 Text-to-Speech - Ekly API

High-quality multilingual TTS from Alibaba’s Qwen 3 model via fal.ai. Supports preset voices, cloned voice embeddings, and 10 languages.


Model id	`fal-qwen3-tts-1.7b`
Media	audio
Submit to	`POST /v1/audio`
Price	5 credits per_generation before discounts
Typical time	about 10.0 s

Request

Required: text
Optional: language, voice_name

{
  "model": "fal-qwen3-tts-1.7b",
  "text": "Welcome back. Your export is ready.",
  "language": "Auto",
  "voice_name": "Vivian"
}

Field	Allowed	Default
`language`	`Auto`, `English`, `Chinese`, `Spanish`, `French`, `German`, `Italian`, `Japanese`, `Korean`, `Portuguese`, `Russian`	`"Auto"`
`voice_name`	`Vivian`, `Serena`, `Ono_Anna`, `Sohee`, `Dylan`, `Eric`, `Ryan`, `Aiden`, `Uncle_Fu`	`"Vivian"`

Reference inputs take asset ids from POST /v1/uploads or an earlier output’s asset_id. Prices and allowed values come from the live catalog; GET /v1/models/fal-qwen3-tts-1.7b is always current.