Built by people who type too much
Loqua started as a tool we made for ourselves. It turns out other people wanted it too
The problem
We're a team of engineers, writers, and researchers. We spend our days composing code comments, Slack messages, client emails, product briefs, and the occasional chapter of fiction. The bottleneck was always the same: our thinking was faster than our typing
Existing voice tools treated dictation as a transcription problem. You speak, it writes down what you said — filler words, false starts, and all. Then you spend time cleaning it up. The net speed gain was marginal
The insight
What we wanted wasn't transcription. We wanted a tool that understood where our words were going and what they needed to look like when they got there. A Slack message doesn't read like an email. A code comment doesn't read like a text to a friend. The output format should match the destination automatically — not because you told it to, but because it read the room
How we got here
Loqua started as an internal prototype in early 2024. A speech model for accuracy. A language model for cleanup and formatting. A multimodal layer that reads the active app and surrounding context. We used it every day, iterated on what felt wrong, and eventually realized we'd built something worth sharing
We launched publicly in late 2024. Since then, Loqua has processed millions of dictations for engineers, founders, writers, researchers, and people who just hate typing on their phones
What we believe
- Speed shouldn't cost accuracy Dictation that needs extensive editing isn't saving anyone time
- Context is the product The same idea needs to land differently in every app. That's not a feature — it's the whole point
- Privacy is non-negotiable Your words are processed ephemerally and never stored. We built it this way from day one, not as an afterthought
- Tools should disappear The best version of Loqua is one you forget is running
The team
We're a small, focused team. AI practitioners who've shipped production ML systems. Engineers who care about latency at the millisecond level. Writers who test every output with their own work. We use Loqua every day — if something feels off, we notice before you do
Get in touch
We'd love to hear from you. Reach us at official@theloqua.ai or visit our contact page