Codesota · Benchmark · MuST-C En-De tst-COMMONHome/Leaderboards/Audio & Speech/Speech Translation/MuST-C En-De tst-COMMON
Unknown

MuST-C En-De tst-COMMON.

Multilingual Speech Translation Corpus built from TED talks. The English-German tst-COMMON split is the de-facto benchmark for end-to-end speech translation. BLEU on tst-COMMON is the primary metric.

Paper ↗Leaderboard ↓
§ 01 · Leaderboard

Results by metric.

Only 3 models on this benchmark
Help build the community leaderboard — submit your model results.
Found a wrong score or missing run?
Use row edits to send a sourced correction into moderation.
Add / edit result ↗Report issue ↗

Bleu

Bleu is the reported evaluation metric for MuST-C En-De tst-COMMON. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Bleuverifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksFix
01SeamlessM4T v2 Large
SeamlessM4T v2 Large, MuST-C En-De tst-COMMON BLEU. seed — verify.
paper37.12026Source ↗Looks wrong?
02Whisper Large v2
Whisper Large-v2 zero-shot speech translation, MuST-C En-De. seed — verify.
paper292026Source ↗Looks wrong?
03Fairseq S2T (MuST-C)
Fairseq S2T conformer baseline on MuST-C En-De tst-COMMON. seed — verify.
paper22.72026Source ↗Looks wrong?
§ 04 · Submit a result

Add to the leaderboard.

← Back to Speech Translation
MuST-C En-De tst-COMMON Leaderboard | CodeSOTA | CodeSOTA