Japanese is one of the hardest languages for automatic speech recognition: three writing systems, no spaces between words, and many homophones. We tested Conference Captioning on real Japanese speech and on a second, reproducible real-time test, and both landed at roughly 95% accuracy or better.
On a real 254-word speech sample. Below 10% is considered good and below 5–6% is top-tier for professional use.
The transcript kept over 95% of the original information, with only 4.75% lost.
In the live test, captions appeared with less than 1.5 seconds of lag, suitable for presentations, classrooms and conferences.
Both tests used the Multilingual speech model, which can switch languages in real time and is not our most accurate model for a single language.
| Type | Count | Examples |
|---|---|---|
| Substitutions (S) | 5 | 温かい→暖かい, 5年→お念, 風→ふう, ピタッと→来たとこ, 未来→見らえ |
| Deletions (D) | 4 | が, です, ね, に |
| Insertions (I) | 2 | あのー, でも |
| Words in reference (N) | 254 |
WER = (5 + 4 + 2) / 254 = 11 / 254 ≈ 4.33%
About a 4.2% word error rate over roughly two minutes of speech, transcribed in real time.
| Reference | Output | Type |
|---|---|---|
| 未来を片づくります | 未来を形作ります | Substitution |
| 上熱を持って | 情熱をもって | Substitution |
| ら心盤 | 羅針盤 | Substitution |
| お忘れないこと | 忘れないこと | Deletion |
| ありがとうございました | ありがとうございまし | Truncation |
The reference text itself contained the odd wording shown on the left, so several “errors” are the model correcting it. None affected meaning.
You can reproduce this test: the speech was generated with ChatGPT using the prompt “Give me a 2 minute inspirational speech written in Japanese for high school students to motivate them to study hard and be passionate about what they do”, played with the Listen feature in Google Translate, and fed into the app.
Our tests measured a 4.33% word error rate (95.25% information preserved) on real speech and about 95.8% accuracy in a live real-time test.
Yes. Captions appeared with less than 1.5 seconds of lag, with no pre-uploaded audio.
Yes. A speaker can talk in Japanese and attendees can read live captions or translations in their own language.