mirrors/icefall
Archived
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'k2-fsa:master' into gigaspeech_recipe

Browse Source
This commit is contained in:
Wang, Guanbo 2022-04-13 18:10:46 -04:00 committed by GitHub
commit 0986b8f9e9
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
6 changed files with 144 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
egs/librispeech/ASR/README.md
View File

@ -9,13 +9,15 @@ for how to run models in this recipe.
There are various folders containing the name `transducer` in this folder.
The following table lists the differences among them.
| | Encoder | Decoder | Comment |
|---------------------------------------|-----------|--------------------|---------------------------------------------------|
| `transducer` | Conformer | LSTM | |
| `transducer_stateless` | Conformer | Embedding + Conv1d | |
| `transducer_lstm` | LSTM | LSTM | |
| `transducer_stateless_multi_datasets` | Conformer | Embedding + Conv1d | Using data from GigaSpeech as extra training data |
| `pruned_transducer_stateless` | Conformer | Embedding + Conv1d | Using k2 pruned RNN-T loss |
| | Encoder | Decoder | Comment |
|---------------------------------------|---------------------|--------------------|---------------------------------------------------|
| `transducer` | Conformer | LSTM | |
| `transducer_stateless` | Conformer | Embedding + Conv1d | |
| `transducer_lstm` | LSTM | LSTM | |
| `transducer_stateless_multi_datasets` | Conformer | Embedding + Conv1d | Using data from GigaSpeech as extra training data |
| `pruned_transducer_stateless` | Conformer | Embedding + Conv1d | Using k2 pruned RNN-T loss |
| `pruned_transducer_stateless2` | Conformer(modified) | Embedding + Conv1d | Using k2 pruned RNN-T loss |
The decoder in `transducer_stateless` is modified from the paper
[Rnn-Transducer with Stateless Prediction Network](https://ieeexplore.ieee.org/document/9054419/).

142
egs/librispeech/ASR/RESULTS.md
View File

@ -1,5 +1,119 @@
## Results
### LibriSpeech BPE training results (Pruned Transducer 2)
[pruned_transducer_stateless2](./pruned_transducer_stateless2)
This is with a reworked version of the conformer encoder, with many changes.
#### Training on fulll librispeech
using commit `34aad74a2c849542dd5f6359c9e6b527e8782fd6`.
See <https://github.com/k2-fsa/icefall/pull/288>
The WERs are:
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|-------------------------------------------------------------------------------|
| greedy search (max sym per frame 1) | 2.62 | 6.37 | --epoch 25 --avg 8 --max-duration 600 |
| fast beam search | 2.61 | 6.17 | --epoch 25 --avg 8 --max-duration 600 --decoding-method fast_beam_search |
| modified beam search | 2.59 | 6.19 | --epoch 25 --avg 8 --max-duration 600 --decoding-method modified_beam_search |
| greedy search (max sym per frame 1) | 2.70 | 6.04 | --epoch 34 --avg 10 --max-duration 600 |
| fast beam search | 2.66 | 6.00 | --epoch 34 --avg 10 --max-duration 600 --decoding-method fast_beam_search |
| greedy search (max sym per frame 1) | 2.62 | 6.03 | --epoch 38 --avg 10 --max-duration 600 |
| fast beam search | 2.57 | 5.95 | --epoch 38 --avg 10 --max-duration 600 --decoding-method fast_beam_search |
The train and decode commands are:
`python3 ./pruned_transducer_stateless2/train.py --exp-dir=pruned_transducer_stateless2/exp --world-size 8 --num-epochs 26 --full-libri 1 --max-duration 300`
and:
`python3 ./pruned_transducer_stateless2/decode.py --exp-dir pruned_transducer_stateless2/exp --epoch 25 --avg 8 --bpe-model ./data/lang_bpe_500/bpe.model --max-duration 600`
The Tensorboard log is at <https://tensorboard.dev/experiment/Xoz0oABMTWewo1slNFXkyA> (apologies, log starts
only from epoch 3).
#### Training on train-clean-100:
Trained with 1 job:
`python3 ./pruned_transducer_stateless2/train.py --exp-dir=pruned_transducer_stateless2/exp_100h_ws1 --world-size 1 --num-epochs 40 --full-libri 0 --max-duration 300`
and decoded with:
`python3 ./pruned_transducer_stateless2/decode.py --exp-dir pruned_transducer_stateless2/exp_100h_ws1 --epoch 19 --avg 8 --bpe-model ./data/lang_bpe_500/bpe.model --max-duration 600`.
The Tensorboard log is at <https://tensorboard.dev/experiment/AhnhooUBRPqTnaggoqo7lg> (learning rate
schedule is not visible due to a since-fixed bug).
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|-------------------------------------------------------|
| greedy search (max sym per frame 1) | 7.12 | 18.42 | --epoch 19 --avg 8 |
| greedy search (max sym per frame 1) | 6.71 | 17.77 | --epoch 29 --avg 8 |
| greedy search (max sym per frame 1) | 6.64 | 17.19 | --epoch 39 --avg 10 |
| fast beam search | 6.58 | 17.27 | --epoch 29 --avg 8 --decoding-method fast_beam_search |
| fast beam search | 6.53 | 16.82 | --epoch 39 --avg 10 --decoding-method fast_beam_search |
Trained with 2 jobs:
`python3 ./pruned_transducer_stateless2/train.py --exp-dir=pruned_transducer_stateless2/exp_100h_ws2 --world-size 2 --num-epochs 40 --full-libri 0 --max-duration 300`
and decoded with:
`python3 ./pruned_transducer_stateless2/decode.py --exp-dir pruned_transducer_stateless2/exp_100h_ws2 --epoch 19 --avg 8 --bpe-model ./data/lang_bpe_500/bpe.model --max-duration 600`.
The Tensorboard log is at <https://tensorboard.dev/experiment/dvOC9wsrSdWrAIdsebJILg/>
(learning rate schedule is not visible due to a since-fixed bug).
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|-----------------------|
| greedy search (max sym per frame 1) | 7.05 | 18.77 | --epoch 19 --avg 8 |
| greedy search (max sym per frame 1) | 6.82 | 18.14 | --epoch 29 --avg 8 |
| greedy search (max sym per frame 1) | 6.81 | 17.66 | --epoch 30 --avg 10 |
Trained with 4 jobs:
`python3 ./pruned_transducer_stateless2/train.py --exp-dir=pruned_transducer_stateless2/exp_100h_ws4 --world-size 4 --num-epochs 40 --full-libri 0 --max-duration 300`
and decoded with:
`python3 ./pruned_transducer_stateless2/decode.py --exp-dir pruned_transducer_stateless2/exp_100h_ws4 --epoch 19 --avg 8 --bpe-model ./data/lang_bpe_500/bpe.model --max-duration 600`.
The Tensorboard log is at <https://tensorboard.dev/experiment/a3T0TyC0R5aLj5bmFbRErA/>
(learning rate schedule is not visible due to a since-fixed bug).
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|-----------------------|
| greedy search (max sym per frame 1) | 7.31 | 19.55 | --epoch 19 --avg 8 |
| greedy search (max sym per frame 1) | 7.08 | 18.59 | --epoch 29 --avg 8 |
| greedy search (max sym per frame 1) | 6.86 | 18.29 | --epoch 30 --avg 10 |
Trained with 1 job, with --use-fp16=True --max-duration=300 i.e. with half-precision
floats (but without increasing max-duration), after merging <https://github.com/k2-fsa/icefall/pull/305>.
Train command was
`python3 ./pruned_transducer_stateless2/train.py --exp-dir=pruned_transducer_stateless2/exp_100h_fp16 --world-size 1 --num-epochs 40 --full-libri 0 --max-duration 300 --use-fp16 True`
The Tensorboard log is at <https://tensorboard.dev/experiment/DAtGG9lpQJCROUDwPNxwpA>
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|-----------------------|
| greedy search (max sym per frame 1) | 7.10 | 18.57 | --epoch 19 --avg 8 |
| greedy search (max sym per frame 1) | 6.81 | 17.84 | --epoch 29 --avg 8 |
| greedy search (max sym per frame 1) | 6.63 | 17.39 | --epoch 30 --avg 10 |
Trained with 1 job, with --use-fp16=True --max-duration=500, i.e. with half-precision
floats and max-duration increased from 300 to 500, after merging <https://github.com/k2-fsa/icefall/pull/305>.
Train command was
`python3 ./pruned_transducer_stateless2/train.py --exp-dir=pruned_transducer_stateless2/exp_100h_fp16 --world-size 1 --num-epochs 40 --full-libri 0 --max-duration 500 --use-fp16 True`
The Tensorboard log is at <https://tensorboard.dev/experiment/Km7QBHYnSLWs4qQnAJWsaA>
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|-----------------------|
| greedy search (max sym per frame 1) | 7.10 | 18.79 | --epoch 19 --avg 8 |
| greedy search (max sym per frame 1) | 6.92 | 18.16 | --epoch 29 --avg 8 |
| greedy search (max sym per frame 1) | 6.89 | 17.75 | --epoch 30 --avg 10 |
### LibriSpeech BPE training results (Pruned Transducer)
Conformer encoder + non-current decoder. The decoder
@ -17,11 +131,15 @@ The WERs are:
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|------------------------------------------|
| greedy search (max sym per frame 1) | 2.62 | 6.37 | --epoch 42, --avg 11, --max-duration 100 |
| greedy search (max sym per frame 2) | 2.62 | 6.37 | --epoch 42, --avg 11, --max-duration 100 |
| greedy search (max sym per frame 3) | 2.62 | 6.37 | --epoch 42, --avg 11, --max-duration 100 |
| modified beam search (beam size 4) | 2.56 | 6.27 | --epoch 42, --avg 11, --max-duration 100 |
| beam search (beam size 4) | 2.57 | 6.27 | --epoch 42, --avg 11, --max-duration 100 |
| greedy search (max sym per frame 1) | 2.62 | 6.37 | --epoch 42 --avg 11 --max-duration 100 |
| greedy search (max sym per frame 2) | 2.62 | 6.37 | --epoch 42 --avg 11 --max-duration 100 |
| greedy search (max sym per frame 3) | 2.62 | 6.37 | --epoch 42 --avg 11 --max-duration 100 |
| modified beam search (beam size 4) | 2.56 | 6.27 | --epoch 42 --avg 11 --max-duration 100 |
| beam search (beam size 4) | 2.57 | 6.27 | --epoch 42 --avg 11 --max-duration 100 |
The decoding time for `test-clean` and `test-other` is given below:
(A V100 GPU with 32 GB RAM is used for decoding. Note: Not all GPU RAM is used during decoding.)
@ -111,7 +229,7 @@ The WERs are
| | test-clean | test-other | comment |
|---------------------------|------------|------------|------------------------------------------|
| greedy search | 2.85 | 6.98 | --epoch 28, --avg 15, --max-duration 100 |
| greedy search | 2.85 | 6.98 | --epoch 28 --avg 15 --max-duration 100 |
The training command for reproducing is given below:
@ -171,8 +289,8 @@ The WERs are
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|------------------------------------------|
| greedy search (max sym per frame 1) | 2.64 | 6.55 | --epoch 39, --avg 15, --max-duration 100 |
| modified beam search (beam size 4) | 2.61 | 6.46 | --epoch 39, --avg 15, --max-duration 100 |
| greedy search (max sym per frame 1) | 2.64 | 6.55 | --epoch 39 --avg 15 --max-duration 100 |
| modified beam search (beam size 4) | 2.61 | 6.46 | --epoch 39 --avg 15 --max-duration 100 |
The training command for reproducing is given below:
@ -241,10 +359,10 @@ The WERs are
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|------------------------------------------|
| greedy search (max sym per frame 1) | 2.67 | 6.67 | --epoch 63, --avg 19, --max-duration 100 |
| greedy search (max sym per frame 2) | 2.67 | 6.67 | --epoch 63, --avg 19, --max-duration 100 |
| greedy search (max sym per frame 3) | 2.67 | 6.67 | --epoch 63, --avg 19, --max-duration 100 |
| modified beam search (beam size 4) | 2.67 | 6.57 | --epoch 63, --avg 19, --max-duration 100 |
| greedy search (max sym per frame 1) | 2.67 | 6.67 | --epoch 63 --avg 19 --max-duration 100 |
| greedy search (max sym per frame 2) | 2.67 | 6.67 | --epoch 63 --avg 19 --max-duration 100 |
| greedy search (max sym per frame 3) | 2.67 | 6.67 | --epoch 63 --avg 19 --max-duration 100 |
| modified beam search (beam size 4) | 2.67 | 6.57 | --epoch 63 --avg 19 --max-duration 100 |
The training command for reproducing is given below:

1
egs/librispeech/ASR/pruned_transducer_stateless/pretrained.py
View File

@ -174,6 +174,7 @@ def main():
# <blk> is defined in local/train_bpe_model.py
params.blank_id = sp.piece_to_id("<blk>")
params.unk_id = sp.piece_to_id("<unk>")
params.vocab_size = sp.get_piece_size()
logging.info(f"{params}")

2
egs/librispeech/ASR/pruned_transducer_stateless2/beam_search.py
View File

@ -89,7 +89,7 @@ def fast_beam_search(
# (shape.NumElements(), 1, joiner_dim)
# fmt: off
current_encoder_out = torch.index_select(
encoder_out[:, t:t + 1, :], 0, shape.row_ids(1)
encoder_out[:, t:t + 1, :], 0, shape.row_ids(1).to(torch.int64)
)
# fmt: on
logits = model.joiner(

4
egs/librispeech/ASR/pruned_transducer_stateless2/scaling.py
View File

@ -367,7 +367,7 @@ class ActivationBalancer(torch.nn.Module):
min_positive: the minimum, per channel, of the proportion of the time
that (x > 0), below which we start to modify the derivatives.
max_positive: the maximum, per channel, of the proportion of the time
that (x > 0), below which we start to modify the derivatives.
that (x > 0), above which we start to modify the derivatives.
max_factor: the maximum factor by which we modify the derivatives for
either the sign constraint or the magnitude constraint;
e.g. with max_factor=0.02, the the derivatives would be multiplied by
@ -413,7 +413,7 @@ class DoubleSwishFunction(torch.autograd.Function):
"""
double_swish(x) = x * torch.sigmoid(x-1)
This is a definition, originally motivated by its close numerical
similarity to swish(swish(x), where swish(x) = x * sigmoid(x).
similarity to swish(swish(x)), where swish(x) = x * sigmoid(x).
Memory-efficient derivative computation:
double_swish(x) = x * s, where s(x) = torch.sigmoid(x-1)

2
icefall/diagnostics.py
View File

@ -111,7 +111,7 @@ def get_diagnostics_for_dim(
options object
sizes_same:
True if all the tensor sizes are the same on this dimension
stats_type: either "abs" or "positive" or "eigs" or "value",
stats_type: either "abs" or "positive" or "eigs" or "value",
imdictates the type of stats we accumulate, abs is mean absolute
value, "positive" is proportion of positive to nonnegative values,
"eigs" is eigenvalues after doing outer product on this dim, sum