Observed arrival · 2026-09-08
Voxbench puts speech-to-text models side by side
Record or upload up to 60 seconds of audio, then compare transcripts from as many as 10 speech-recognition models.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
Field notes
Voxbench accepts a short recording or upload, applies optional vocabulary and language settings, and sends the same take to selected transcription providers. The interface lists 10 models, including systems from Google, OpenAI, ElevenLabs, AssemblyAI, Deepgram, Microsoft, and NVIDIA. An optional reference transcript measures substituted, missing, and extra words, while provider-specific notes show which models accept vocabulary hints and which do not.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
⊠LoginAccess appeared gated
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
ƒJavaScriptBrowser-side code central
One card from the complete issue