Supported Languages
60+ input languages, 17 output languages. No manual selection needed.
Input Languages (Speech Recognition)
VoxInk uses Soniox for speech recognition, which auto-detects your language from 60+ languages. You don't need to select your input language — just start speaking.
Supported input languages include but are not limited to: English, Chinese (Mandarin, Cantonese), Japanese, Korean, Spanish, French, German, Portuguese, Italian, Russian, Arabic, Hindi, Thai, Vietnamese, Indonesian, Malay, Turkish, Polish, Dutch, Swedish, Norwegian, Danish, Finnish, Czech, Greek, Hebrew, Romanian, Hungarian, Ukrainian, Bengali, Tamil, Telugu, Urdu, Swahili, Filipino, Burmese, Khmer, Lao, and many more.
Code-Switching
VoxInk handles mixed-language speech. If you naturally switch between languages mid-sentence (e.g., English with Chinese phrases, or Spanish with English terms), the transcription captures both languages accurately.
Output Languages (Translate & Listen)
For Translate mode and Listen mode captions, VoxInk supports these 17 output languages:
How Languages Work Per Mode
Direct Mode
Input language is auto-detected. Output is the same language you spoke in. No translation.
Polish Mode
Input language is auto-detected. Output is the same language, but cleaned up by AI. Works in any language.
Translate Mode
Input: any of 60+ languages (auto-detected). Output: your chosen target language from the 17 above.
Listen Mode
Captions are translated in real-time to any of the 17 output languages. Each listener can choose their own language.
Related
- Translate Mode — speak in one language, output in another
- Listen Mode — real-time translated captions
- Getting Started — set up VoxInk in 2 minutes