LibriSpeech librispeech Leaderboard
Auto-discovered from papers reporting LibriSpeech (WER). · Metric: WER (lower is better)
| # | Model | WER | Paper |
|---|---|---|---|
| 1 | Decoder-only Conformer with Modality-aware Sparse Mixtures of Experts for ASR | 2.80 | — |
| 2 | Language model integration based on memory control for sequence to sequence speech recognition | 3.70 | — |
| 3 | Language Model Integration Based On Memory Control For Sequence To Sequence Speech Recognition | 3.70 | — |
| 4 | Whisfusion: Parallel ASR Decoding via a Diffusion Transformer | 8.30 | — |
| 5 | Segaug: Ctc-aligned Segmented Augmentation For Robust Rnn-transducer Based Speech Recognition | 12.50 | — |
| 6 | SegAug: CTC-Aligned Segmented Augmentation For Robust RNN-Transducer Based Speech Recognition | 12.50 | — |
| 7 | Towards Data-free and Training-free Compression for Speech Foundation Models Using Parameter Clustering | 27.73 | — |