Skip to main content

Learn Learning paths Courses

Voice AI Agents — a pipeline that listens, looks things up and speaks

Wire listening through speaking in a single CPU pod and measure the latency

고급 · Lessons 30 · Lab 10

Curriculum

Audio basics — PCM, frames, resampling

VAD and turn-taking — end-of-speech detection and barge-in

Streaming ASR — partial results, final results, WER

Streaming LLM responses and time to first token

RAG for spoken queries — rewriting, refusal thresholds, citation checks

Agent workflow — tools, confirmation, retries, handoff

Streaming TTS — sentence splitting and time to first audio

Latency budget — running the pipeline overlapped

Evaluation and observability — quality, failure rate, SLO, gates

Safety — injection and personal data arriving by voice

Reference docs