Voice, as a practice.
Mira is an open-source, end-to-end encrypted voice training companion for trans people. Live feedback from a full DSP pipeline — pitch, weight, resonance, prosody, volume — that never leaves your device.
Try the prototype →“Good morning. How's everything going today?”
Built on the work of trans voice educators.
The five-slider concept model, the vowel placement exercises, and the coaching framework come from Seattle Voice Lab. The exercise progressions and the exploration-first framing are shaped by Trans Voice Lessons.
Partnership discussions are pending. Mira is not affiliated with or endorsed by either program. If you are either of the above and want Mira to stop using your framework, open a GitHub issue and we will comply immediately.
What we will and won't do.
On-device analysis
The DSP pipeline runs in your browser. Audio is never uploaded in plaintext.
E2E-encrypted storage
Recordings are encrypted with keys derived from your password. We never see them.
No match scores
You are the expert on your voice. Mira shows approaching / in-range / exploring — never a percentage.
No red in the UI
Red carries clinical and failure weight. This is a practice space, not a pass/fail test.