Deepgram Nova-3 vs Reson8
Deepgram Nova-3 is ahead on time to first text and final-text delay, and Reson8 on the Pipecat Dataset. The Ocular Dataset is too close to call.
Last updated Methodology by Dileep Chagam
Accuracy vs speed
Each dot is one model. Bottom-left is best: fewer errors and a faster final transcript.
Which to pick
Pick Deepgram Nova-3 for showing words while the caller is still talking (1.34s sooner) and quick turn-taking once the caller stops (132ms sooner). Pick Reson8 for accuracy on clean, scripted audio (1.37 points lower). For accuracy on real phone conversations, the gap is inside the margin of error, so neither is ahead.
The four metrics, side by side
- Deepgram Nova-3
- 3.46%12th
- Reson8
- 2.10%4th
Reson8, 1.37 points lower
- Deepgram Nova-3
- 4.39%11th
- Reson8
- 2.93%1st
Too close to call: inside the margin of error
- Deepgram Nova-3
- 1.06s7th
- Reson8
- 2.40s13th
Deepgram Nova-3, 1.34s sooner
- Deepgram Nova-3
- 103ms4th
- Reson8
- 235ms12th
Deepgram Nova-3, 132ms sooner
Lower is better on every metric. The small figure is each model's place in the whole field. What each metric measures.
- WER · Word error rate
- Share of words the model got wrong: replaced, missing or added. Lower is better.
- Pipecat Dataset
- 1,000 short public clips from Pipecat's STT benchmark dataset.
- Ocular Dataset
- 8 recordings from 4 real phone conversations, licensed from Ocular.
- TTFT · Time to first text
- How long after the audio starts streaming until the first words appear.
- TTFS · Final-text delay
- How long after the caller stops until the full transcript arrives. A voice agent waits for this before it replies.
First text and final text, at three percentiles
Deepgram Nova-3 shows first text after 2.07s and final text after 155ms, and Reson8 shows first text after 10.20s and final text after 398ms, on a bad turn (P95) across the same streamed turns.
| Measure | Model | P50 | P90 | P95 |
|---|---|---|---|---|
| Time to first text | Deepgram Nova-3 | 1.06s | 1.10s | 2.07s |
| Reson8 | 2.40s | 7.97s | 10.20s | |
| Final-text delay | Deepgram Nova-3 | 103ms | 143ms | 155ms |
| Reson8 | 235ms | 357ms | 398ms |
- P50 / P90 / P95 · Typical, slow and slowest turns
- Half of turns are faster than P50. Only 1 in 10 is slower than P90, and 1 in 20 slower than P95.
How each ranks among all 17 models
Place among the 17 models on each metric. Lower is better on all four.
| Metric | Nova-3 | Reson8 |
|---|---|---|
| Pipecat Dataset WER | 3.46%12th of 17 | 2.10%4th of 17 |
| Ocular Dataset WER | 4.39%11th of 17 | 2.93%1st of 17 |
| Time to first text | 1.06s7th of 15 | 2.40s13th of 15 |
| Final-text delay | 103ms4th of 17 | 235ms12th of 17 |
How much audio each model was tested on
| Model | Pipecat clips | Ocular recordings | Turns timed | Failed |
|---|---|---|---|---|
| Deepgram Nova-3 | 1,000 | 8 | 202 | 0 |
| Reson8 | 1,000 | 8 | 201 | 0 |
Only clips with usable text count, so the two totals can differ. How WER is scored.
Related comparisons
- Deepgram Nova-3 vs GPT-4o Mini Transcribe
- Deepgram Nova-3 vs Smallest Pulse
- Deepgram Nova-3 vs GPT-4o Transcribe
- Deepgram Nova-3 vs Deepgram Flux English
- Deepgram Nova-3 vs Inworld STT-1
- Deepgram Nova-3 vs Soniox STT-RT v5
- Deepgram Nova-3 vs Speechmatics Linden
- Deepgram Nova-3 vs Cartesia Ink 2
- Deepgram Nova-3 vs GPT Realtime Whisper
- Deepgram Nova-3 vs Google Chirp 3
- Deepgram Nova-3 vs Deepgram Flux Multilingual
- Deepgram Nova-3 vs AssemblyAI Universal 3.6 Pro
- Deepgram Nova-3 vs Gradium
- Reson8 vs GPT Realtime Whisper
- Reson8 vs Google Chirp 3
- Reson8 vs Cartesia Ink 2
- Reson8 vs AssemblyAI Universal 3.6 Pro
- Reson8 vs Speechmatics Linden
- Reson8 vs Soniox STT-RT v5
- Reson8 vs Inworld STT-1
- Reson8 vs GPT-4o Mini Transcribe
- Reson8 vs Smallest Pulse
- Reson8 vs GPT-4o Transcribe
- Reson8 vs Deepgram Flux English
- Reson8 vs Deepgram Flux Multilingual
- Reson8 vs Gradium
Frequently asked questions
Which is more accurate, Deepgram Nova-3 or Reson8?
Reson8 is more accurate on the public clips. On the 1,000 public Pipecat clips, Deepgram Nova-3 scores 3.46% and Reson8 2.10%, so Reson8 makes fewer errors by 1.37 points. On the eight licensed Ocular recordings, Deepgram Nova-3 scores 4.39% and Reson8 2.93%, with likely ranges of 3.42 to 5.25% and 2.29 to 3.68%. The ranges overlap, so the difference is too close to call. Word error rate counts incorrect, missing, and extra words, so lower is better.
Is Deepgram Nova-3 faster than Reson8?
Yes. Deepgram Nova-3 finishes the transcript 132ms sooner. Deepgram Nova-3 has the final transcript 103ms after the speaker stops and Reson8 235ms, at the median. Both are measured on the same 206 streamed turns.
Is Deepgram Nova-3 or Reson8 better for voice agents?
Deepgram Nova-3 finishes the transcript 132ms sooner, and the two are too close to call on real conversations. A voice agent cannot reply until the final transcript arrives, and the Ocular recordings are real conversations.
Do Deepgram Nova-3 and Reson8 stream text while the speaker is talking?
Yes, both return partial text before the speaker stops. Deepgram Nova-3 shows first text after 1.06s and Reson8 after 2.40s, at the median, so Deepgram Nova-3 is 1.34s sooner.