Inworld STT-1 vs Reson8
Inworld STT-1 is ahead on time to first text and final-text delay, and Reson8 on the Ocular Dataset. The Pipecat 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 Inworld STT-1 for showing words while the caller is still talking (1.34s sooner) and quick turn-taking once the caller stops (194ms sooner). Pick Reson8 for accuracy on real phone conversations (7.67 points lower). For accuracy on clean, scripted audio, the gap is inside the margin of error, so neither is ahead.
The four metrics, side by side
- Inworld STT-1
- 2.72%10th
- Reson8
- 2.10%4th
Too close to call: inside the margin of error
- Inworld STT-1
- 10.60%16th
- Reson8
- 2.93%1st
Reson8, 7.67 points lower
- Inworld STT-1
- 1.06s6th
- Reson8
- 2.40s13th
Inworld STT-1, 1.34s sooner
- Inworld STT-1
- 41ms2nd
- Reson8
- 235ms12th
Inworld STT-1, 194ms 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
Inworld STT-1 shows first text after 1.07s and final text after 65ms, 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 | Inworld STT-1 | 1.06s | 1.07s | 1.07s |
| Reson8 | 2.40s | 7.97s | 10.20s | |
| Final-text delay | Inworld STT-1 | 41ms | 60ms | 65ms |
| 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 | STT-1 | Reson8 |
|---|---|---|
| Pipecat Dataset WER | 2.72%10th of 17 | 2.10%4th of 17 |
| Ocular Dataset WER | 10.60%16th of 17 | 2.93%1st of 17 |
| Time to first text | 1.06s6th of 15 | 2.40s13th of 15 |
| Final-text delay | 41ms2nd of 17 | 235ms12th of 17 |
How much audio each model was tested on
| Model | Pipecat clips | Ocular recordings | Turns timed | Failed |
|---|---|---|---|---|
| Inworld STT-1 | 1,000 | 8 | 206 | 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
- Inworld STT-1 vs Soniox STT-RT v5
- Inworld STT-1 vs Speechmatics Linden
- Inworld STT-1 vs Cartesia Ink 2
- Inworld STT-1 vs GPT Realtime Whisper
- Inworld STT-1 vs GPT-4o Mini Transcribe
- Inworld STT-1 vs Google Chirp 3
- Inworld STT-1 vs Deepgram Nova-3
- Inworld STT-1 vs Smallest Pulse
- Inworld STT-1 vs AssemblyAI Universal 3.6 Pro
- Inworld STT-1 vs GPT-4o Transcribe
- Inworld STT-1 vs Deepgram Flux English
- Inworld STT-1 vs Deepgram Flux Multilingual
- Inworld STT-1 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 GPT-4o Mini Transcribe
- Reson8 vs Deepgram Nova-3
- 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, Inworld STT-1 or Reson8?
Reson8 is more accurate on real conversations. On the 1,000 public Pipecat clips, Inworld STT-1 scores 2.72% and Reson8 2.10%, with likely ranges of 1.67 to 4.68% and 1.86 to 2.35%. The ranges overlap, so the difference is too close to call. On the eight licensed Ocular recordings, Inworld STT-1 scores 10.60% and Reson8 2.93%, so Reson8 makes fewer errors by 7.67 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
Is Inworld STT-1 faster than Reson8?
Yes. Inworld STT-1 finishes the transcript 194ms sooner. Inworld STT-1 has the final transcript 41ms after the speaker stops and Reson8 235ms, at the median. Both are measured on the same 206 streamed turns.
Is Inworld STT-1 or Reson8 better for voice agents?
It is a trade-off. Reson8 makes fewer errors on real conversations, while Inworld STT-1 finishes the transcript 194ms sooner. A voice agent cannot reply until the final transcript arrives, and the Ocular recordings are real conversations.
Do Inworld STT-1 and Reson8 stream text while the speaker is talking?
Yes, both return partial text before the speaker stops. Inworld STT-1 shows first text after 1.06s and Reson8 after 2.40s, at the median, so Inworld STT-1 is 1.34s sooner.