Coval published benchmarks showing Inworld AI STT 1 improved its WER to 4.2% while holding a 63ms median TTFS, pushing Cartesia Ink-2 off the frontier.
Public source
Publisher name
Public post
This week in benchmarks: TTS latency keeps dropping! fluxions Vui keeps the fastest-TTS spot after last week's mid-week drop, settling at a 51ms median TTFA. Its WER is…
Company
Coval
Simulation & Evaluation for AI Voice & Chat Agents.
- Industry
- Technology, Information and Internet
- Location
- San Francisco, US
- Company size
- 11–50 employees
About Coval
Coval accelerates AI agent development with automated testing for chat, voice, and other objective-oriented systems. Many engineering teams are racing to market with AI agents, but slow manual testing processes are holding them back. Teams currently play whack-a-mole just to discover that fixing one issue introduces another. At Coval, we use automated simulation and evaluation techniques inspired by the autonomous vehicle industry to boost test coverage, speed up development, and validate consistent performance.
See moreLatest activity
Latest activity from Coval
30 signals
Discover more
Similar signals
Similar public activity from other companies.
Research & Knowledge
Gradium
Gradium was benchmarked by Voice Arena on 10 TTS models on latency, landing as the lowest p50 at 231 ms with the tightest spread across p25, p50, and p75.
Research & Knowledge
Besimple AI
Besimple AI published how Thinking Machines Lab's Inkling speech model was fine-tuned on 1, 25, and 100 hours of proprietary data containing alphanumeric entities.
Research & Knowledge
Voice Arena
Voice Arena measured 231 ms to first audio at the median with a p25-p75 spread of about 7 ms.
Research & Knowledge
Speechmatics
Speechmatics benchmarked Melia 1 for switch-point accuracy across Arabic, Mandarin, and Tamil paired with English against Soniox, Amazon, and AssemblyAI.
Research & Knowledge
Shunya Labs