whatmodel.sh / stt / Universal-Streaming (English)

Universal-Streaming (English)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Universal-Streaming (English) is a closed-weights speech-to-text model from AssemblyAI, released June 2025. Licence: proprietary.

Catalogue identity notes

AssemblyAI Universal-Streaming, English-only realtime model (API speech_model universal-streaming-english); introduced 2 June 2025 (AssemblyAI blog 'Introducing Universal-Streaming'). Hosted alias: the vendor may have updated weights since launch; the Pipecat run dates from the repository's first results (by 2026-06-04, commit 9e24957 shows the row with the same TTFS figures) and was re-scored on 2026-10-01.

58.3±3.3#44 of 162 speech-to-text modelsfrom 1 benchmark result across 1 source* thinly tested, treat the rank as provisional

Score within the Open ASR English short-form WER (March 2026) comparison set only.

$ how it compares

The speech-to-text models ranked either side of Universal-Streaming (English), out of 176 with a score, then more from AssemblyAI.

#modelscorewhereRAMreleased
51Parakeet TDT 1.1B NVIDIA59±2local~5.8GB2024-01
52Voxtral Mini 3B (2507) Mistral58±1api $0.0010/audio min +local~12GB2025-07
53Universal-Streaming (English) AssemblyAI58±3api $0.0025/audio min2025-06
54Parakeet RNNT 1.1B NVIDIA58±2local~5.6GB2024-01
55Chirp 2 Google58±3api price unknown
more from AssemblyAI
4Universal-3.6 Pro Realtime AssemblyAI70±3api $0.0075/audio min2026-09
8Universal-3.5 Pro AssemblyAI67±3api $0.0035/audio min2026-07
17Universal-3 Pro AssemblyAI63±2comparison only2026-02

see all speech-to-text models · find the ones that fit your machine

$ what it's good at

other tests1 not in a task · 96% of its score
testresultvs the others in its rundifficultycountswhy
Pipecat STT Benchmark · pooled semantic WER (1000 smart-turn samples, Sonnet 5.5 judge)2.4% ✎▃·▃▅·▅▅·▃▅██ beats 4/16avg 4.8%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 96%typical

$ where to run it

Through an API

api · $0.0025/audio min · www.assemblyai.com ↗ · price sourceUSD per input audio minute: AssemblyAI pricing page, Realtime Speech-to-Text API, 'Universal-Streaming ... High-quality realtime model for English only transcription ... $0.15 /hr' (÷60), billed on session duration. Product: AssemblyAI Realtime STT API. API model: universal-streaming-english. Price basis: published list rate, per-hour to per-minute scaling only. Exclusions: add-ons. Currency: USD. AssemblyAI official pricing page, observed 2026-10-09.

$ how the score was worked out · Pipecat STT Benchmark · seman… 96%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Pipecat STT Benchmark · semantic WER results table (pipecat-ai/stt-benchmark README)1▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 96%BSD-2-Clause
prior · pull toward the middle▇ 4%
weighted average= 58.3

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json