File size: 1,163 Bytes
9fe5087
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
# k2_zipformer2_english_v1

- Zipformer2 recipe derived from : https://github.com/k2-fsa/icefall/tree/master/egs/librispeech/ASR/zipformer

- Training data : CommonVoice, VoxPopuli (speed-perturb 3x):
   - cca 1500 + 3x500 hours of training data

- Output text/symbols include:
   - TrueCase capitalization
   - punctuation `[,.?!]` as standalone tokens

## Config:
```
--num-epochs 20 \
--base-lr 0.04 \
\
--causal 1 \
--use-transducer 1 \
--use-ctc 0 \
\
--num-encoder-layers 2,2,2,2,2,2 \
--feedforward-dim 512,768,768,768,768,768 \
--encoder-dim 192,256,256,256,256,256 \
--encoder-unmasked-dim 192,192,192,192,192,192 \
```

## Results

| ID | System | cv-dev | cv-test | vp-dev | vp-test | Comment |
|---|-------------------------|--------|---------|--------|---------|-------------------------------|
| A | small (24M) | 18.57 | 21.98 | 13.66 | 13.26 | ep20,avg4 |

- non-streaming results from [decode.py](https://github.com/BUTSpeechFIT/k2_streaming_training/blob/main/training/zipformer/decode.py)
- cv = CommonVoice, vp = VoxPopuli
- exported the model averaging : ep20,avg4

## Note
- Not the best possible results, this model is for integration tests