Choose Where Your Audio Is Processed
Overview
Each KizunaAI engine runs a separate deployment per region, and your audio is processed in the region you pick. The Soniox engine offers the United States, the European Union and Japan; the Qwen engine offers Singapore and Beijing. The defaults are the United States and Singapore.
1Open the region setting
Open Settings, switch to Advanced with the button at the top right of the panel, and choose the Provider tab. Region is right below the AI service provider.
2Pick a region
With the Soniox engine, choose United States, European Union or 日本 (Japan); with the Qwen engine, Singapore or 北京 (Beijing), for example the region your organization requires for its data. The choice applies from the next session.
3Add your voice in the new region
With the Soniox engine, custom voices belong to one region, and each region remembers its own voice choice. If you use My voice, add it again after switching; see Choose a voice or clone your own. The Qwen engine’s voices are not tied to a region: its presets and live cloning work the same in Singapore and Beijing.
Tips
- The region cannot be changed while a session is starting or running. Stop the session first.
- Everything else, including languages, vocabulary and the session background, stays the same when you switch region.
- Beijing may answer with noticeably higher latency from networks outside mainland China; if it does, try Singapore.
Frequently Asked Questions
Which region should I choose?
If your organization has rules about where data is processed, pick the region they require. Otherwise the default works for everyone: the United States for Soniox, Singapore for Qwen. Beijing is there for networks in mainland China.
Where is my audio processed if I never change this?
In the engine’s default region: the United States for Soniox, Singapore for Qwen.
Troubleshooting
Still stuck? Send us feedback from the account menu, or write to support@kizuna.ai. You can also ask in GitHub Discussions.