# How Gradient Labs uses Deepgram

**300ms+**: “300ms+ reduction in end-to-end transcription latency”

As published on [deepgram.com](https://deepgram.com/customers/gradient-labs-delivers-human-quality-financial-voice-support). Captured by usedby on 2026-10-09.

- Company: [Gradient Labs](https://www.usedby.ai/companies/gradient-labs.md)
- Tool: [Deepgram](https://www.usedby.ai/tools/deepgram.md)

## What the story says

Gradient Labs builds AI voice support agents for regulated financial institutions and uses Deepgram Nova-3 as the primary real-time speech-to-text engine in a multi-provider transcription pipeline. It also uses Deepgram's EU hosting endpoints to keep processing local.

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

> 300ms+ Latency Reduction: By prioritizing Deepgram's results and moving to EU hosting, end-to-end latency dropped significantly, making conversations feel immediate and engaging.

## What usedby checked

We compared the story with its live page on 2026-10-09.

- Checked: the figure 300ms+ is printed word for word on the page, near the name of Gradient Labs.
- Checked: the passage quoted above is copied word for word from the page, near the name of Gradient Labs.
- Checked: Gradient Labs uses Deepgram. Confirmed line. Latest check across sources: 2026-10-09.
- Not checked: the result itself. We quote it; we did not measure it.

---
Source: https://www.usedby.ai/case-studies/gradient-labs-deepgram · How we check: https://www.usedby.ai/methodology