# How Retell uses Cartesia

**2-3x**: “Cartesia’s Time-to-First-Audio (TTFA) is 2-3x faster than the next best provider on Retell’s platform.”

As published on [cartesia.ai](https://www.cartesia.ai/customers/retell), November 2025. Captured by usedby on 2026-10-08.

- Company: [Retell](https://www.usedby.ai/companies/retell.md)
- Tool: [Cartesia](https://www.usedby.ai/tools/cartesia.md)

## What the story says

Retell AI, a no-code platform for LLM-powered voice agents used in call centers, uses Cartesia's text-to-speech as a primary voice provider and as an automatic failover. Retell customers can enable Cartesia voices on their agents with a single API call.

Summary written by usedby from the source page, in English. The figures are those of Cartesia and Retell, not ours.

> Cartesia’s Time-to-First-Audio (TTFA) is 2-3x faster than the next best provider on Retell’s platform. For voice agents, this translates to conversations that feel natural rather than stilted - critical for maintaining engagement in high-stakes calls.

- **\<0.1%**: “Cartesia maintains a \<0.1% error rate on Retell’s platform.”
- **99.9%**: “Enterprise reliability 99.9% uptime.”

> It’s incredibly difficult to achieve the rare blend of latency, reliability, accuracy, and voice quality. Cartesia managed to do all four, which is why customers adopted them faster than any other voice model we’ve introduced.
>
> Zexia Zhang, CTO of Retell AI

## What usedby checked

We compared the story with its live page on 2026-10-08.

- Checked: the figure 2-3x is printed word for word on the page, near the name of Retell.
- Checked: the passage quoted above is copied word for word from the page, near the name of Retell.
- Checked: the publication date is read from the page’s own metadata, never guessed.
- Checked: Retell uses Cartesia. Confirmed line. Latest check across sources: 2026-10-08.
- Not checked: the result itself. We quote it; we did not measure it.

---
Source: https://www.usedby.ai/case-studies/retell-ai-cartesia · How we check: https://www.usedby.ai/methodology