csukuangfj commited on
Commit
093a289
1 Parent(s): be5bd12

Update readme.

Browse files
Files changed (1) hide show
  1. README.md +4 -3
README.md CHANGED
@@ -19,18 +19,19 @@ git lfs pull
19
 
20
  **Catuion**: You have to run `git lfs pull`. Otherwise, you will be SAD later.
21
 
22
- The model in this repo is trained using the commit `TODO`.
23
 
24
  You can use
25
 
26
  ```
27
  git clone https://github.com/k2-fsa/icefall
28
  cd icefall
29
- git checkout TODO
30
  ```
31
  to download `icefall`.
32
 
33
- You can find the model information by visiting <https://github.com/k2-fsa/icefall/blob/TODO/egs/librispeech/ASR/transducer_stateless_multi_datasets/train.py#L198>.
 
34
 
35
  In short, the encoder is a Conformer model with 8 heads, 12 encoder layers, 512-dim attention, 2048-dim feedforward;
36
  the decoder contains a 1024-dim embedding layer and a Conv1d with kernel size 2.
 
19
 
20
  **Catuion**: You have to run `git lfs pull`. Otherwise, you will be SAD later.
21
 
22
+ The model in this repo is trained using the commit `2332ba312d7ce72f08c7bac1e3312f7e3dd722dc`.
23
 
24
  You can use
25
 
26
  ```
27
  git clone https://github.com/k2-fsa/icefall
28
  cd icefall
29
+ git checkout 2332ba312d7ce72f08c7bac1e3312f7e3dd722dc
30
  ```
31
  to download `icefall`.
32
 
33
+ You can find the model information by visiting
34
+ <https://github.com/k2-fsa/icefall/blob/2332ba312d7ce72f08c7bac1e3312f7e3dd722dc/egs/librispeech/ASR/transducer_stateless_multi_datasets/train.py#L218>
35
 
36
  In short, the encoder is a Conformer model with 8 heads, 12 encoder layers, 512-dim attention, 2048-dim feedforward;
37
  the decoder contains a 1024-dim embedding layer and a Conv1d with kernel size 2.