Deepgram opens a Sydney region for in-country audio processing
- Source
- x.com
- Date
The Deepgram Australia endpoint is now generally available. 🇦🇺 https://t.co/mS6c4HNQGX runs in Sydney, with your audio, transcripts, and speech output processed and stored entirely within Australia. Same API and models. Moving over is a base URL change, no self-hosting required. Blog →
Context
Deepgram's Australia endpoint provides managed speech processing in Sydney. The company says that by default storage and for audio, transcripts and generated speech happen in Australia, and that existing API keys and SDK integrations continue to work after pointing at the new endpoint. Its launch article explains the scope.
That scope has boundaries. For the Voice , a managed third-party handles the reasoning step and processes requests outside Australia, and operational metadata and billing are processed in the US. Deepgram says setting mip_opt_out=true (opting out of model improvement) extends residency to all processing and access. Regional speech hosting therefore does not automatically make every stage of an agent domestic.
- inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
- AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
- LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
Checking sign-in…
Loading comments…