saybench

Level FSource

Benchmark the full voice-AI pipeline on your own audio — STT (batch + streaming), LLM TTFT, speech-to-speech, and TTS. WER by failure mode, keyterm recall, real-time-paced latency, cost columns, a CI regression gate, and an MCP server for coding agents.

⭐ 1AI GenerationGoRepo ↗Updated 2026-09-09

Installation

View source repo →

Related servers