Grok TTS tops voice humanness benchmark

// 1h agoBENCHMARK RESULT

Grok TTS tops voice humanness benchmark

Vapi’s blind-voted Humanness Index currently ranks xAI’s Grok TTS as its most human-sounding voice model, with Grok’s streaming variant also near the top. The benchmark compares cloned voices across models so listeners judge model quality, not polished vendor demos.

// ANALYSIS

Grok TTS looks like a serious voice-agent contender, but the win matters most if xAI can pair naturalness with low-latency, stable developer APIs.

–Blind voting is a useful counterweight to cherry-picked demo reels, especially in TTS where tiny artifacts break trust.
–Grok TTS leads on perceived humanness, while its streaming version trades some score for faster response time.
–Vapi’s index puts xAI directly against ElevenLabs, MiniMax, Canopy, and Inworld in a way voice-agent builders can actually compare.
–For production agents, the next question is not just realism; it is latency, cost, language coverage, and reliability under phone-quality audio.

// TAGS

grok-ttsxaivapittsspeechvoice-agentbenchmarkevaluationagent

DISCOVERED

1h ago

2026-06-18

PUBLISHED

2h ago

2026-06-18

RELEVANCE

8/ 10

AUTHOR

elonmusk

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

INFRA2h ago

Kong Insomnia manages and tests modern API protocols

Insomnia, maintained by Kong, is a comprehensive open-source API client that supports testing and debugging for GraphQL, REST, WebSockets, SSE, and gRPC. It provides developers with versatile environment management and flexible storage options, including Cloud, Local, and Git integrations, making it a robust tool for API development workflows.

OPEN SOURCE2h ago

Kilo Code open-source agentic platform builds code faster

Kilo Code is an open-source agentic engineering platform that functions as an all-in-one assistant for coding, debugging, planning, and task orchestration. It integrates seamlessly into various environments like VS Code, JetBrains IDEs, and the CLI while supporting over 500 AI models.

BENCHMARK2h ago

GPT-5.5 Instant matches thinking models in health

OpenAI says GPT-5.5 Instant is now on par with its frontier Thinking models for health-related questions, after feedback from hundreds of physicians across 60 countries, 49 languages, and 26 specialties. The update matters because ChatGPT is now handling health and wellness questions from more than 230 million people each week.