Tal como se publicó en deepgram.com. Verificado por usedby el 9 oct 2026.
Qué pasó
Telnyx runs Deepgram Flux for real-time speech-to-text on its own GPUs at its telephony points of presence, so call audio is transcribed inside its network. The transcripts feed its call control and agent orchestration layer for turn-taking, LLM reasoning and text-to-speech in voice AI calls.
Resumen escrito por usedby a partir de la página de origen, en inglés. Las cifras son de Deepgram y de Telnyx, no nuestras.
Because Deepgram Flux is running at the edge, inside the media plane, Telnyx can deliver sub-second end-to-end latency under load while preserving the reliability and observability of its carrier network.
We chose Deepgram Flux for its high accuracy and consistent real-time performance, running directly on Telnyx-managed GPUs at the edge so our voice experiences feel instant; latency is constrained by physics, not external APIs.
Lo que dice la historia, y lo que verificamos
Lo que comparamos con la página.
- El pasaje citado arribaVerificadoCopiado palabra por palabra de la página, cerca del nombre de Telnyx.
- Telnyx usa DeepgramVerificadoLínea de nivel Confirmado.
- El resultado en síNo verificadoLo citamos; no lo medimos.






