90%+
“90%+ of conversations handled end-to-end by AI agents”
As published on deepgram.com. Captured by usedby on Oct 2, 2026.
What happened
Klubi uses Deepgram's Nova-3 streaming speech-to-text as the recognition layer for AI voice agents that handle pre-sales contact, qualification and handoff for its consórcio group purchasing product in Brazil. The transcription also supports service and post-sales workflows, with transcripts feeding monitoring and QA.
Summary written by usedby from the source page, in English. The figures are those of Deepgram and Klubi, not ours.
- ~80k“~80k total monthly customer interactions across pre-sales, post-sales, and support”
- ~30k“~30k monthly AI-led voice interactions”
Klubi rebuilt its funnel around AI voice, anchored by Deepgram Speech-to-Text (STT), to increase conversion, expand coverage, and improve consistency across thousands of interactions.
We evaluated several speech-to-text providers, but Deepgram stood out immediately for Brazilian Portuguese. Other solutions struggled with domain-specific vocabulary—confusing car model names like 'Ônix' with common words like 'ônibus' [bus]. Deepgram got it right, even in noisy environments like traffic or subway calls. That accuracy was non-negotiable for us.
What the story claims, and what we checked
We compared the story with its live page on Oct 3, 2026.
- The figure: 90%+CheckedPrinted word for word on the page, near the name of Klubi.
- The passage quoted aboveCheckedCopied word for word from the page, near the name of Klubi.
- The numbers in our summaryCheckedEach one is printed on the page.
- Klubi uses DeepgramCheckedConfirmed line. Latest check across sources: Oct 2, 2026.
- The result itselfNot checkedWe quote it; we did not measure it.
More stories
Same company, same tool or same industry.
