From 6a8b649e56671ad9f9e5fa3ae13a2a1c177411e8 Mon Sep 17 00:00:00 2001 From: Desh Raj Date: Mon, 13 Feb 2023 02:53:28 -0500 Subject: [PATCH] Add small streaming Zipformer transducer model (#903) --- egs/librispeech/ASR/RESULTS.md | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/egs/librispeech/ASR/RESULTS.md b/egs/librispeech/ASR/RESULTS.md index 1a894498e..ecb84eb01 100644 --- a/egs/librispeech/ASR/RESULTS.md +++ b/egs/librispeech/ASR/RESULTS.md @@ -77,6 +77,18 @@ for m in greedy_search modified_beam_search fast_beam_search; do done ``` +#### Smaller model + +A smaller model (~20M params) is also available with configuration based on [this comment](https://github.com/k2-fsa/icefall/pull/745#issuecomment-1405282740). The WERs are: + +| decoding method | chunk size | test-clean | test-other | comment | decoding mode | +|----------------------|------------|------------|------------|---------------------|----------------------| +| greedy search | 320ms | 3.94 | 9.79 | --epoch 30 --avg 9 | simulated streaming | +| modified beam search | 320ms | 3.88 | 9.53 | --epoch 30 --avg 9 | simulated streaming | + +You can find a pretrained model, training logs, decoding logs, and decoding +results at: + ### zipformer_mmi (zipformer with mmi loss)