Integrations
Use local speech tools without sending source audio to the web.
Local speech API
Trusted tools on the same computer connect to the desktop app's local gateway at 127.0.0.1:1117 — an OpenAI-compatible API authenticated with the Bearer key shown in the app. Responses use the shared JSON output shape.
POST /v1/audio/transcriptions # speech → text POST /v1/audio/alignments # known text → word timing POST /v1/text/segmentations # raw text → sentences GET /v1/models # available models POST /mcp # MCP over HTTP
License activation
After purchase, copy your license key into the desktop app to activate devices. Activated devices can then be managed from your account.
Downloads
The download page always points to the current desktop build, with Windows support clearly marked as coming soon.
Payments
Purchases are completed through a secure payment flow. EdgeSpeak does not store card numbers.