NeuroDynamicLive
How the speech services fit together
Three parts, three jobs. How do they connect?
Speech processing, API access and the desktop app each do one thing. Here’s how they connect, and where to find the guides to set up your own.
The goalPrivate, flexible dictationUse our hosted service, or your own server.
Where it’s atLiveLexVoca and the APIs are in use today.
What it costsFree to startLexVoca is free. The APIs are paid per request.
How it works
Follow the arrows. Each part has a guide below.
LexVoca on your PCRecords your speech and types the returned text on Windows.
A speech serverTurns a recording into text. Use ours, or run your own.
The public APIPay-per-request speech, narration and approved voice cloning for apps and agents.
Results so far
What each part is for.
Your choiceHosted, or your own compatible speech server
Per requestAPI access for apps and agents, no subscription
SeparateAPI payments and LexVoca accounts are kept apart
What we learned
- Speech needs capacity. A server needs enough memory for the model and the recordings it processes.
- Keep jobs separate. Dictation, paid API access and accounts each have their own records, so one can change without breaking the others.
- Shared graphics cards fill up. Running several models on one card needs planning. The guide below shows how.
Build your own
The guides, in order.
Start at the top. Each guide shows how hard it is and how long it takes.