... using your own Deepgram key (note: Cloudflare currently supports linear16 encoding only.) Post-call analytics Capture transcripts and recordings once a call finishes and run analysis with LLMs.  Read more about how to set up post-call workflows . Cartesia Sonic-3 emotive voices Support for Cartesia

... Cartesia's Sonic-3 is now available on Layercode. Released last week, Sonic-3 is the fastest and most expressive production-ready TTS model: Incredible low-latency: 90ms model latency, 190ms end-to-end makes Sonic-3 ideal for real-time voice AI agents and one of the fastest TTS models available. Natural emotion and expressiveness: Sonic-3 handles laughter, excitement, and empathy. You can use SSML tags for emotions, for example: <emotion value="excited" />Oh wow, Valentine ... ... Intelligent text handling: Sonic-3 understands text like a human, reading acronyms, homophones, abbreviations, numbers, dates, and addresses correctly based on context. Multilingual: Supports 42+ languages with native pronunciation. Full language support coming soon to Layercode. Voice cloning: Use Cartesia

... features Latency breakdowns: View latency breakdowns for all agent sessions (i.e. total latency, user endpoint latency, agent backend response latency, TTS generation latency). Just click into any session > 'View latency breakdown' under a session's 'Average latency' Cartesia Sonic-3 STT:  Added preview support for Cartesia

... Kokoro 82 M have developed in lockstep: in blind tests, most listeners can t reliably separate them from the incumbents. Broadly, today's models fall into two distinct categories that serve fundamentally different purposes: Real-time models like Cartesia Sonic, ElevenLabs Flash, and Hexgrad Kokoro prioritize streaming audio generation, producing speech as text arrives rather than waiting for complete sentences. These models excel in conversational AI where low-latency makes the difference between natural dialogue and awkward pauses. These models ... ... working on one of the voice models we ve tested for this post or one we haven t we d love to explore an integration. Comparing today s leading text-to-speech voice models Beyond the real-time VS