Deepgram
Overview
Deepgram's developer-first speech recognition API platform, led by the Nova ASR line (current Nova-3, plus Flux for conversational streaming), known for low-latency streaming transcription, per-minute pricing and diarization.
Field checks and source notes
When sources disagree on a field, every reading is listed here with its own citation. We do not take a midpoint, invent a missing figure, or decide which side is right — follow the citation and judge for your case.
These field-level notes are recorded verbatim in Chinese in the source data. Open the Chinese version
Plain-language benchmarks
Practical facts
| Context length | Not disclosed |
|---|---|
| Max output | Not disclosed |
| Parameters | Not disclosed |
| Input | Not disclosed |
| Output price | Not disclosed |
| CNY | Not disclosed |
| Free tier | New accounts receive $200 of credit. |
| Output speed | Not disclosed |
| Release date | |
| License | Not disclosed |
Show technical data
Technical benchmark table
| Benchmark | Score | Max / unit | Source type | Source | Test date | Conditions |
|---|---|---|---|---|---|---|
| ASR word error rate asr-wer | 5.2% | percent / 100 | Independent test | AA-WER 指数:约 8 小时、三数据集按时长加权平均,批量转写;越低越好 |
| Version | Release date | Status | Openness | Parameters | What changed | Detail | |
|---|---|---|---|---|---|---|---|
| Nova-3Showing version | Active | Closed source | Not disclosed | Released February 2025 for batch and streaming transcription; vendor launch figures cite ~6.84% streaming WER and ~5.3% median batch WER (self-reported), with medical variants and the later Flux conversational model. | Detail | ||
| Nova-2 | Superseded | Closed source | Not disclosed | Previous flagship superseded by Nova-3; exact release date not verified in this pass and left blank. | Detail |
Last updated: · Sources on detail page