Papers › Librispeech Transducer Model with Internal Language Model Prior Correction

Librispeech Transducer Model with Internal Language Model Prior Correction

7 Apr 2021arXiv:2104.03006archive 2025-07-28

Albert Zeyer, André Merboldt, Wilfried Michel, Ralf Schlüter, Hermann Ney

We present our transducer model on Librispeech. We study variants to include an external language model (LM) with shallow fusion and subtract an estimated internal LM. This is justified by a Bayesian interpretation where the transducer model prior is given by the estimated internal LM. The subtraction of the internal LM gives us over 14% relative improvement over normal shallow fusion. Our transducer has a separate probability distribution for the non-blank labels which allows for easier combination with the external LM, and easier estimation of the internal LM. We additionally take care of including the end-of-sentence (EOS) probability of the external LM in the last blank probability which further improves the performance. All our code and setups are published.

PaperPDFCode

Code

rwth-i6/returnn officialtfNOASSERTION report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Language ModelingLanguage ModellingSentenceSpeech Recognitionmodel

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Recognition LibriSpeech test-clean LSTM Transducer Word Error Rate (WER) 2.23 #33 of 64 Archive leaderboard report
Speech Recognition LibriSpeech test-other LSTM Transducer Word Error Rate (WER) 5.6 #34 of 53 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections