During your event
Live interpreter
- Timing
- About 2.4 seconds
- Translation quality
- About the same as human translations in the three languages tested
- Speech recognition
- 1st in Arabic and Korean
Accuracy
How accurate is Baian?
See how Baian compares with professional translators, interpreters and other AI. Every number has a source.
Start with people01 · Against translators
They read the text. Baian only heard it.
Baian still scored higher in Hausa, Tamil, Swahili and Urdu. Its results were about the same in 21 more languages.
Professional translatorsRead the text
BaianHeard the audio
Baian’s subtitles were preferred:
- Hausa75%preferred
- Tamil69%preferred
- Swahili68%preferred
- Urdu67%preferred
The line marks 50%: an even split.
01 · Against translators, live
What about live translation?
About the same quality as professional translations in Arabic, Hindi and Korean, while working live from speech.
How often Baian was preferred
- Arabic58%preferred
- Hindi65.5%preferred
- Korean61.5%preferred
The line marks 50%: an even split.
02 · Against interpreters
How far behind the speaker?
The speaker’s own words appear first. The translation follows about 2.4 seconds behind the speech.
Baian vs human interpretersTranslation follows in about 2.4 seconds.Seconds behind the speaker. Lower is better.
1.0 sOriginal captions2.4 sBaian translation2.7 sInterpreters
03 · Against other AI
First place in seven languages.
On independent tests that compare OpenAI, Google, ElevenLabs, Microsoft and others.
Live speech: transcription rankingsErrors in words, or characters for Korean and Japanese. Lower is better.
-
ArabicLive
Baian5.1%1stSpeechmatics Enhanced6.5%2nd -
KoreanLive
Baian2.9%1stGoogle Gemini 3.5 Live3.8%2nd
Recorded speech: transcription rankingsErrors in words, or characters for Korean and Japanese. Lower is better.
-
SpanishRecorded
Baian3.3%1stOpenAI4.3%2nd -
RussianRecorded
Baian4.7%1stGoogle Gemini 2.5 Flash6.4%2nd -
HindiRecorded
Baian1stRank only -
FrenchRecorded
Baian4.3%Joint 1stAssemblyAI4.6%Joint 1st -
JapaneseRecorded
Baian2.51%Joint 1stElevenLabs2.52%Joint 1st
By product
Full results by product.
After recording
Caption tool
- Timing
- Processes your uploaded recording
- Translation quality
- Ahead in Hausa, Tamil, Swahili and Urdu; about the same in 21 others
- Speech recognition
- 1st in five languages
Transparency
Most interpretation services don’t show their tests.
| Service | Published accuracy evidence |
|---|---|
| Baian | Results on named public tests, with dates and methods. See them ↗ |
| Interprefy | Its own private test only (“up to 98%”), scored with its own method. Source ↗ |
| Wordly | A 98% to 99% claim with no named test set. Source ↗ |
| KUDO | A private May 2023 test that KUDO itself calls out of date. Source ↗ |
| Interactio | No accuracy figures. Source ↗ |
| Maestra | No accuracy figures. Source ↗ |
| Timekettle | An “up to 98%” claim with no named test set. Source ↗ |
| Microsoft Teams Interpreter | No figures; Microsoft says accuracy isn’t guaranteed. Source ↗ |
Questions
Questions about accuracy.
Is Baian more accurate than a human translator?
In the recorded-speech test, Baian scored above professional translations in Hausa, Swahili, Tamil and Urdu. The judge found no clear difference in 21 other languages. Live translation had about the same overall quality as professional translations in Arabic, Hindi and Korean. These results do not apply to every recording or language.
Is Baian faster than a human interpreter?
Our estimate for Baian’s first translated text is 2.4 seconds behind the speaker. A separate study measured European Parliament interpreters at a median 2.69 seconds. Different recordings and methods mean this is a reference comparison, not proof Baian is always faster.
How does Baian compare with OpenAI, Google and ElevenLabs?
On independent speech tests Baian ranked first in seven languages: live Arabic and Korean, and recorded Spanish, Russian and Hindi, with joint first places in French and Japanese.
How were the tests run?
On public test sets that other companies are measured on, with dates and methods published below. In the translation tests, Baian worked from spoken audio and the professional translators worked from written text. The human-interpreter timing figure comes from a separate published study.
Hear it with your own event audio.
Bring a short recording, your languages and your audience size. We’ll show you the listener experience.
Arrange a demoSources and methodsEvery number on this page, with its test and date
- Live speech recognition (AssemblyAI real-time benchmark)AssemblyAI runs 25 August to 23 September 2026; Baian runs 13 to 15 October 2026
1st in Arabic (5.1%) and Korean (2.9%), 2nd in Russian and Danish. Fewer errors than all three OpenAI real-time models in 13 of the 14 languages we publish.
AssemblyAI real-time benchmarkMethod
FLEURS: short Wikipedia sentences read aloud. Measures how many words come out wrong.
AssemblyAI published scores for 14 real-time engines. We ran Baian’s live interpreter on the same test set.
We joined clips into 15-minute recordings with 1.5-second gaps and played them into Baian’s live interpreter in real time, like a microphone, with automatic language detection. AssemblyAI streamed each clip separately with the language set in advance. Scoring ignores punctuation, capital letters, Arabic vowel marks and how numbers are written. Our charts show 13 of AssemblyAI’s 14 engines; the full table is on AssemblyAI’s site.
- Recorded speech recognition (Verli)Verli July 2026; Baian October 2026
1st in Spanish, Russian and Hindi; joint 1st in French and Japanese.
Verli speech-to-text accuracy, July 2026Method
200 FLEURS sentences per language. Measures words wrong.
Verli tested 15 or 16 engines per language. We ran Baian’s caption tool on the same test set.
Verli tested each engine as it was in July 2026 on its own 200-sentence FLEURS sample. We used the same test set and scoring rule on a different 200 sentences. Verli’s scoring rule splits Hindi words at vowel signs, which roughly halves everyone’s error rate, so we publish Hindi as a rank only.
- Baian against professional translators (FLoRes)October 2026
Caption Tool scored above the translators in Hausa, Swahili, Tamil and Urdu (preferred on 67% to 75% of sentences). Baian’s live interpreter had about the same overall quality in Arabic, Hindi and Korean.
FLoRes-200 professional translationsMethod
148 English sentences that also have professional human translations. Measures translation quality with Google’s MetricX-24 automatic judge.
Baian ran the test. The human translations come from the public FLoRes-200 set.
Baian worked from the spoken audio; the translators worked from text. We compared scores sentence by sentence and counted a result only when the 95% range did not cross zero and two runs agreed. “About the same” means the test found no clear difference; it does not prove the translations were identical.
- English speech to Egyptian Arabic (WMT26)15 October 2026
2nd of 30 systems, and preferred over the professional post-edited translation on 91 of 133 clips (68%).
WMT26 general translation testMethod
133 English YouTube clips, the speech test from the 2026 Conference on Machine Translation.
Baian scored every published system’s output, and its own, with Google’s MetricX-24 automatic judge.
Baian received the same automatic transcript as the other systems. WMT’s own human raters rank systems differently, so we always describe this as an automatic score.
- Urdu speech recognition (Qwen3.5-Omni paper)Paper April 2026; Baian October 2026
13.2% of words wrong, the lowest in the comparison. OpenAI GPT-4o Transcribe got 19.7%.
Qwen3.5-Omni paper, April 2026 (arXiv)Method
FLEURS Urdu recordings. Measures words wrong.
The Qwen3.5-Omni authors published scores for several systems; we ran the Caption Tool on the same test set.
Scored with the standard Whisper rule: lower case, punctuation removed.
- Live delay13 to 15 October 2026
Median 2.1 seconds across 53 recordings in 15 languages (1.3 to 2.7 seconds per recording).
Method
How long after a sentence ends its finished translation appears.
Baian, on its production Live service.
Measured on Baian’s production Live service from 13 to 15 October 2026, from the end of each spoken sentence to the finished English line in our captured listener feed. Earlier words in a long sentence wait longer than this. Spoken audio arrives a little after the text and was not measured.
- How quickly words appear13 to 15 October 2026
Estimated medians: the speaker’s words appear in their own language about 1.0 seconds after they are spoken, and the first translation covering a word about 2.4 seconds after it.
Method
Estimated from our captured listener feed for the same 53 recordings. The capture is incomplete, so some early previews are missing, and previews can still change before the sentence is finished. Times are measured to the feed, not to a phone screen.
- European Parliament interpreters2019
Median 2.69 seconds behind the speaker, across 180 speeches.