How accurate is Inworld STT-1?

Inworld STT-1 transcribes the 1,000 public Pipecat clips at 2.72% word error rate and the eight licensed Ocular recordings at 10.60%. Word error rate counts incorrect, missing, and extra words against the reference transcript, so lower is better.

How fast is Inworld STT-1 in a live conversation?

First text arrives after 1.06s at the median, and the transcript is final 41ms after the speaker stops. Both are measured on the same 206 streamed turns as every other model, and the page shows P90 and P95 as well, which is what a caller meets on a bad turn.

How was Inworld STT-1 measured?

Every model runs the same two datasets under the same conditions: 1,000 public clips from Pipecat's STT benchmark dataset for accuracy at scale, and eight licensed conversation recordings from Ocular for real speech. Latency comes from 206 streamed turns, using each provider's own supported finalization contract. Inworld STT-1 returned usable text for 1,000 of the public clips.

Bench

Book a demo

Inworld STT-1

Every modelInworld

Accuracy vs. speed

Lower and further left is better · Nemotron 3.5 Transcribe doesn't support streaming, so it is not shown

Best quadrantBest picks

Final-text delay P50 (ms)

Overall WER (%)

Inworld STT-1 is 2nd of 18 on final-text delay and 18th of 19 on the Ocular Dataset.

5.44% Overall WER (17th) · 2.72% Pipecat WER (12th) · 10.60% Ocular WER (18th) · 1.06s First text (7th) · 41ms Final text (2nd)Likely range, 95% confidence: Pipecat WER 1.67 to 4.68%, Ocular WER 3.85 to 22.76%Published price $0.15 per hour of streamed audio. On-demand rate; $0.10/hr on paid plans. Source, checked Oct 8, 2026.

Last updated Sep 16, 2026Methodology by Dileep Chagam

Inworld STT-1 against every other model

Lower is better on every measure. Hover a column for that model's full numbers.

NVIDIA Nemotron 3.5 Transcribe send no text until the speaker stops, so TTFT is not comparable.

Glossary: WER, TTFT, TTFS, P50

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.
Likely range · 95% range: Where the model's true score probably falls, given how much audio was tested. Overlapping ranges are too close to call.
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.
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.

Where Inworld STT-1 struggled

Extra words are its biggest source of error on both real calls and clean clips. Our guide explains how word error rate is measured.

72%

of its errors on the Ocular calls are words no one said.

Median across all 19 models: 14%

Mix of errors

Real calls (Ocular Dataset)

Clean clips (Pipecat Dataset)

TTFT and TTFS at P50, P90 and P95

Timed on the same 206 streamed Ocular turns as every other model. P90 and P95 show what a caller meets on a slow turn.

TTFS, final text

How long after the caller stops until the full transcript arrives. A voice agent waits for this before it replies.

20ms50ms100ms200ms500ms1s2s

Typical, P50: 41ms; 2nd fastest of 18
Slow, P90: 60ms; Fastest of 18
Slowest, P95: 65ms; Fastest of 18

TTFT, first text

How long after the audio starts streaming until the first words appear.

200ms500ms1s2s5s10s20s

Typical, P50: 1.06s; 7th fastest of 16
Slow, P90: 1.07s; 3rd fastest of 16
Slowest, P95: 1.07s; 2nd fastest of 16

Every other model's typical turn, on the same axis.

1,000 Pipecat clips · 8 Ocular recordings · 206 turns timed · 0 failed requests

Inworld STT-1 FAQ

Where does Inworld STT-1 place in the benchmark?

Inworld STT-1 places highest on final-text delay, 2nd of 18 at 41ms. Its weakest placing is the Ocular Dataset, 18th of 19 at 10.60%.

Inworld STT-1 places highest on final-text delay, 2nd of 18 at 41ms. Its weakest placing is the Ocular Dataset, 18th of 19 at 10.60%.
Inworld STT-1 transcribes the 1,000 public Pipecat clips at 2.72% word error rate and the eight licensed Ocular recordings at 10.60%. Word error rate counts incorrect, missing, and extra words against the reference transcript, so lower is better.
First text arrives after 1.06s at the median, and the transcript is final 41ms after the speaker stops. Both are measured on the same 206 streamed turns as every other model, and the page shows P90 and P95 as well, which is what a caller meets on a bad turn.
Every model runs the same two datasets under the same conditions: 1,000 public clips from Pipecat's STT benchmark dataset for accuracy at scale, and eight licensed conversation recordings from Ocular for real speech. Latency comes from 206 streamed turns, using each provider's own supported finalization contract. Inworld STT-1 returned usable text for 1,000 of the public clips.

Building on Inworld? Test your agent the same way.

Cekura runs these simulated calls on your agent and scores every one.

Book a demo Start free