Personal project · voice · 2026

Loquai

A voice studio in the browser: turn text into speech, manage a library of voices and transcribe audio.

Role
Concept and development
Period
2026
Stack
PythonFastAPIReactSQLiteMistral Voxtral

A — The problem

People who produce audio content jump between tools: one for synthesis, one for voices, one for transcripts, each with its own account and limits.

B — What I built

Loquai brings it all into one web studio: a text editor with voice generation, a voice library, transcription and job history, structured to support multiple users and quotas.

C — How it works

  1. 01

    Write

    The text to be read aloud.

  2. 02

    Pick a voice

    From your personal library.

  3. 03

    Generate and listen

    The file stays in your history.

  4. 04

    Transcribe

    The reverse path, from audio to text.

D — The system

A simplified map of the components: not the technical documentation, but it shows how the parts talk to each other.

E — Why it works

One place

Synthesis, voices and transcription together.

Multi-user ready

Separate accounts, quotas and history.

Selective overview: code and configuration remain private.