Deepgram speech-to-text is now natively available on @togethercompute.
Run your STT, LLM, and TTS on one platform — co-located so audio doesn't have to leave the environment to get transcribed. One API, one bill, Deepgram quality on every transcript.
Built for the audio that actually matters: contact centers, healthcare, financial services — not just clean recordings.
Call (847) 851-4323 to try a live voice agent running on the pipeline.
Link to blog:
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.
inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
Why it matters
Voice pipelines normally cross a network boundary between transcription and inferenceRunning a trained model to get answers — the phase where AI is actually used, as opposed to trained.Full definition →. Running Deepgram STT next to the model on Together removes that hop and collapses the API surface and billing to one platform.
