Local project storage
Project content and editing state are stored through the desktop app’s native data layer.
Desktop speech studio
Turn words into voice.
Bring every voice back to words.
Create natural speech, transcribe live or recorded audio, and keep every project in one focused desktop workspace.
Windows macOS Version 1.0.2
LiveAudio ready
144Languages & dialects
From idea to finished audio
A simple workflow.
Powerful results.
Move from raw text or audio to a finished project without switching between disconnected tools.
- 01
Start with content
Type a script, import audio, record a microphone, or capture desktop sound.
- 02
Shape the result
Choose voices, tune delivery, mark effects, or refine a live transcript.
- 03
Save and export
Keep the complete project and export audio or a transcript when it is ready.
01 / 02
Multi-voice editing, right where you write
Text to Speech
Direct every line,
not just the document.
Build natural speech with document-wide engine controls and precise settings for the passages that need a different voice, pace, emphasis, or effect.
- 01
Voice where it matters
Apply different voices and delivery settings to selected words or passages.
- 02
Make pronunciation intentional
Reuse dictionary entries for names, abbreviations, dates, and specialist terms.
- 03
Export without guesswork
Create WAV, MP3, OGG, or WebM audio, with conversion when the engine needs it.
Speech to Text
Capture the moment.
Keep the meaning.
Transcribe microphones, desktop audio, or recordings into a living canvas. See live text while you work, then edit and save it as a project.
- 01
Live transcription
Watch new speech arrive in the canvas during microphone or desktop capture.
- 02
Your choice of recognition
Use user-managed Whisper models with automatic GPU preference and CPU fallback.
- 03
Know who said what
Separate speakers in supported workflows with configurable speaker detection.
02 / 02
Download and manage recognition models
Built around your workflow
Your projects.
Your providers. Your control.
Voxnara keeps project work focused on the desktop and makes connected speech services an explicit choice.
Bring your own provider
Connect supported speech providers with your own credentials and see only the capabilities they offer.
No silent fallback
Offline recognition does not quietly send audio to a cloud service when a local model is unavailable.
Inside the workspace
Built for the work
between input and output.
Ready when you are
Your next project
already has a voice.
Download Voxnara and bring text, voice, and transcription into one desktop workspace.