How to Create AI Judge Metrics You Can Trust
How to use human review in Coval to calibrate your LLM-as-a-judge metrics: label conversations, check agreement, read disagreements, fix the prompt, and re-test.
Blog
Coval's observations on developing and shipping AI voice agents
We raised $28M. Here's what's next for us.
TTS Latency 2026: The Production Speed Metric for Voice AI
More reading
Best TTS Providers 2026: Why Vendor Benchmarks Lie
Best TTS providers in 2026: ElevenLabs v3, Cartesia Sonic 3, OpenAI Realtime, Deepgram Aura-2, Vapi Voices. Latency, pricing, voice cloning. See the guide.
Vapi Review 2026: Composer, Evals & Vapi Voices
Vapi review for 2026: Composer builder, Vapi Voices, Monitoring, Evals, Simulations, GPT-5 + Realtime, and post-Series B pricing. Read the guide.
ElevenLabs Voice Cloning Review 2026: v3, Scribe & Agents
ElevenLabs voice cloning in 2026: IVC vs PVC requirements, Eleven v3 vs Flash v2.5, Scribe v2 STT, ElevenAgents, and the May pricing reset. Read the guide.
CI/CD for Voice AI Agents: The 2026 Pipeline Guide
Voice agents are software. CI/CD pipelines with automated eval gates are how mature teams ship them. Patterns, tools, and pipeline design.
Voice AI Testing Build vs. Buy: When Internal Tools Hit a Wall
Voice AI testing build vs. buy: most teams build first, then hit the wall. The framework, the math, and what good buying actually looks like.
Voice AI Regression Testing: Escape Whack-a-Mole
Voice AI regression testing turns whack-a-mole shipping into safe iteration. Methodology, scenario design, and tooling for production teams.
Benchmarks
Independent TTS benchmarks from Coval: how Gradium, ElevenLabs, Cartesia, Rime, and OpenAI compare on time to first audio and word error rate.
Voice AI in Production: Why Agents That Pass Testing Still Break in the Wild
Why voice AI in production handles 60-70% of calls when staging suggested 95%. The conditions test sets miss, and how Coval teams close the gap.
Voice Observability for Voice AI in Production
Voice observability tells teams whether AI agents are working in production. Coval's complete guide to behavioral monitoring at scale. Read the guide.