CekuraBench
Speech to textText to speech
Book a demo
All benchmarks›Vendors›Cartesia

Cartesia

1 model across speech to text.

Last updated Nov 3, 2026 · Methodology by Dileep Chagam

Building on Cartesia?

Test your agent the same way.

Book a demoStart free

Speech to text

Full results →
Ink 22.63% WER4th of 19

Other vendors

  • OpenAI
  • Google
  • xAI
  • Deepgram
  • Amazon
  • AssemblyAI
  • Boson AI
  • Deepslate
  • Microsoft
  • Phonic
  • Fish Audio
  • Gradium
  • Inworld
  • NVIDIA
  • Reson8
  • Rime
  • Smallest
  • Soniox
  • Speechmatics
Cekura

Public benchmark results for voice agent platforms, realtime speech models, and speech recognition systems.

Benchmarks
All benchmarks
Speech to speechAgent performanceTurn benchReal human calls
Speech to textText to speech
Voice platformsAgent performanceVoice quality
Explore
Compare S2S modelsCompare STT modelsCompare platformsVendorsExperimentsAbout the benchmarkGovernance and changelogSource repositoryDataset license
Cekura
WebsiteDocumentationBlogContact
© 2026 CekuraTermsPrivacy
Cartesia · 1 model benchmarked