Voxta docs
Voxta ServerServicesSpeech-to-Text

Vosk

Vosk is an open-source, offline speech-to-text engine for Voxta that runs on modest hardware. Lighter but less accurate than Whisper.

Vosk is an open-source offline STT engine. Built to run on modest hardware including Android and Raspberry Pi. Lower accuracy than cloud STT or Whisper-based local STT, but very lightweight.

Setup

Add the service

Manage Modules → Add Modules → Vosk → Install.

Pick a language model

Vosk requires a language-specific model. The config provides a model picker — pick one for your language. Voxta downloads it on first use.

SenseVoice

SenseVoice is a local speech-to-text engine for Voxta that also detects emotion and audio events like laughter, music and applause.

Whisper Live

Open-source neural STT based on Whisper. Strong accuracy, multilingual.

On this page

SetupAdd the servicePick a language model