Cartesia Ink 2 vs Deepgram Nova-3
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner.
- Ink 22.39% · 123ms
- Nova-33.46% · 103ms
Cartesia Ink 2 is ahead on both datasets together and the Pipecat Dataset, and Deepgram Nova-3 on time to first text and final-text delay. The Ocular Dataset is too close to call.
Last updated Methodology by Dileep Chagam
Who is better at what
- Overall WER2.63%3.78%
- Pipecat Dataset WER2.39%3.46%
- Ocular Dataset WER3.09%4.39%
- Time to first text1.51s1.06s
- Final-text delay123ms103ms
Lower is better on every measure. Too close to call means the likely ranges overlap. For its voices, see our notes on Cartesia text to speech.
Ink 2 vs Nova-3 FAQ
Which is more accurate, Cartesia Ink 2 or Deepgram Nova-3?
Cartesia Ink 2 is more accurate on both datasets. On the 1,000 public Pipecat clips, Cartesia Ink 2 scores 2.63% and Deepgram Nova-3 3.78%, so Cartesia Ink 2 makes fewer errors by 1.16 points. On the eight licensed Ocular recordings, Cartesia Ink 2 scores 2.39% and Deepgram Nova-3 3.46%, so Cartesia Ink 2 makes fewer errors by 1.08 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
Cartesia Ink 2 is more accurate on both datasets. On the 1,000 public Pipecat clips, Cartesia Ink 2 scores 2.63% and Deepgram Nova-3 3.78%, so Cartesia Ink 2 makes fewer errors by 1.16 points. On the eight licensed Ocular recordings, Cartesia Ink 2 scores 2.39% and Deepgram Nova-3 3.46%, so Cartesia Ink 2 makes fewer errors by 1.08 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
No. Deepgram Nova-3 finishes the transcript 447ms sooner. Cartesia Ink 2 has the final transcript 1.51s after the speaker stops and Deepgram Nova-3 1.06s, at the median. Both are measured on the same 206 streamed turns.
It is a trade-off. Cartesia Ink 2 makes fewer errors on real conversations, while Deepgram Nova-3 finishes the transcript 447ms sooner. A voice agent cannot reply until the final transcript arrives, and the Ocular recordings are real conversations.
Yes, both return partial text before the speaker stops. Cartesia Ink 2 shows first text after 3.09% and Deepgram Nova-3 after 4.39%, at the median.
Test your voice agent the same way.
Cekura runs these simulated calls on your agent and scores every one.