Type
Spec
Rate
Synthesis
per 10,000 characters
$0.14
StepAudio 3 TTS and Qwen Audio 3.0 TTS side by side: specs, capabilities, parameters, and live pricing from the EmpirioLabs catalog.
StepFun 
Alibaba Cloud Audio GenerationStepAudio 3 TTS from StepFun and Qwen Audio 3.0 TTS from Alibaba Cloud are both in the Audio Generation category, available through the EmpirioLabs API.
| Spec | StepAudio 3 TTS | Qwen Audio 3.0 TTS |
|---|---|---|
| Model ID | stepaudio-3-tts | qwen-audio-3-0-tts |
| Author | StepFun | Alibaba Cloud |
| Released | Sep 9, 2026 | Jul 20, 2026 |
| Input | Text | Text |
| Output | Audio | Audio |
| Region | International | Singapore |
| Endpoints | POST /v1/audio/speech POST /v1/audio/speech:stream | POST /v1/audio/speech POST /v1/audio/speech:stream GET /v1/voices |
| Capabilities |
Live rates from the EmpirioLabs catalog for both models. Each rate card always matches what the API charges.
Request parameters each model lists on EmpirioLabs, with the allowed values or range where one applies. A dash marks a parameter the model does not list.
| Parameter | StepAudio 3 TTS | Qwen Audio 3.0 TTS |
|---|---|---|
| input | max 1000 | max 20000 |
| voice | lively-girl, vibrant-youth, soft-spoken-gentleman, magnetic-v... Default: lively-girl | loongjameszhao, loongolivialin, loonglunawang, loongnorahu, l... Default: loongjameszhao |
| instruction | max 500 | max 128 |
| speed | 0.5 to 2 Default: 1 | 0.5 to 2 Default: 1 |
| volume | 0.1 to 2 Default: 1 | 0 to 100 Default: 50 |
| language | en, zh, ja, ko, fr, es Default: en | - |
| response_format | mp3, wav, flac, opus, pcm Default: mp3 | mp3, wav, pcm, opus Default: mp3 |
| sample_rate | 8000, 16000, 22050, 24000 Default: 24000 | 8000, 16000, 22050, 24000, 44100, 48000 Default: 24000 |
| markdown_filter | boolean Default: false | - |
| pronunciation_map | string | - |
| model_tier | - | plus, flash Default: plus |
| voice_audio_url | - | string |
| voice_id | - | string |
| pitch | - | 0.5 to 2 Default: 1 |
| language_hints | - | zh, en |
| seed | - | 0 to 65535 Default: 0 |
Maximum input is 1,000 characters. Content in parentheses is treated as delivery direction and is not spoken. Use instruction for global emotion or style guidance, up to 500 characters. Twelve system voices are available. Recognized languages are Chinese, English, Japanese, Korean, French, and Spanish, with languages other than Chinese and English in preview. Output formats are mp3, wav, flac, opus, and pcm at 8000, 16000, 22050, or 24000 Hz. Wav is returned as a complete file rather than streamed. This model does not support voice_label.
StepAudio 3 TTS serves speech through POST /v1/audio/speech and returns playable audio. Send the text to speak as input with the model id stepaudio-3-tts. Both models are called through the same endpoint. Set model to stepaudio-3-tts or qwen-audio-3-0-tts. Get an API key from the EmpirioLabs dashboard.
curl https://api.empiriolabs.ai/v1/audio/speech \
-H "Authorization: Bearer $EMPIRIOLABS_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "stepaudio-3-tts",
"input": "Welcome to EmpirioLabs. Your build just finished."
}' \
--output speech.mp3Prompt tests and comparison runs that cover both models.
StepAudio 3 TTS is the newer release (Sep 9, 2026); Qwen Audio 3.0 TTS was released on Jul 20, 2026. Capabilities listed only for StepAudio 3 TTS: contextual tts, voice control. Capabilities listed only for Qwen Audio 3.0 TTS: expressive prosody, emotion control, voice tags, dialect, low latency, voice cloning. The specs, rate cards, and parameters on this page come from the live EmpirioLabs catalog.
StepAudio 3 TTS is the newer release (Sep 9, 2026); Qwen Audio 3.0 TTS was released on Jul 20, 2026.
The two models are priced on different units or tiers, so compare the live rate cards on this page. They always match what the API charges.
Yes. Both models use POST /v1/audio/speech with the same EmpirioLabs API key, so most integrations switch by changing the model value. Check the parameters table on this page first if you rely on a model-specific parameter.
Yes. Open each model in the EmpirioLabs playground and run the same prompt on both, with the parameters the API exposes.
Check out our pricing or reach out if you want your own model deployed on our stack.
