|
|
@ -1,41 +1,3 @@
|
|
|
|
# https://yaml.org/type/float.html
|
|
|
|
|
|
|
|
data:
|
|
|
|
|
|
|
|
train_manifest: data/manifest.train
|
|
|
|
|
|
|
|
dev_manifest: data/manifest.dev
|
|
|
|
|
|
|
|
test_manifest: data/manifest.test-clean
|
|
|
|
|
|
|
|
min_input_len: 0.5 # seconds
|
|
|
|
|
|
|
|
max_input_len: 30.0 # seconds
|
|
|
|
|
|
|
|
min_output_len: 0.0 # tokens
|
|
|
|
|
|
|
|
max_output_len: 400.0 # tokens
|
|
|
|
|
|
|
|
min_output_input_ratio: 0.05
|
|
|
|
|
|
|
|
max_output_input_ratio: 100.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
collator:
|
|
|
|
|
|
|
|
vocab_filepath: data/lang_char/vocab.txt
|
|
|
|
|
|
|
|
unit_type: 'spm'
|
|
|
|
|
|
|
|
spm_model_prefix: 'data/lang_char/bpe_unigram_5000'
|
|
|
|
|
|
|
|
mean_std_filepath: ""
|
|
|
|
|
|
|
|
augmentation_config: conf/preprocess.yaml
|
|
|
|
|
|
|
|
batch_size: 16
|
|
|
|
|
|
|
|
raw_wav: True # use raw_wav or kaldi feature
|
|
|
|
|
|
|
|
spectrum_type: fbank #linear, mfcc, fbank
|
|
|
|
|
|
|
|
feat_dim: 80
|
|
|
|
|
|
|
|
delta_delta: False
|
|
|
|
|
|
|
|
dither: 1.0
|
|
|
|
|
|
|
|
target_sample_rate: 16000
|
|
|
|
|
|
|
|
max_freq: None
|
|
|
|
|
|
|
|
n_fft: None
|
|
|
|
|
|
|
|
stride_ms: 10.0
|
|
|
|
|
|
|
|
window_ms: 25.0
|
|
|
|
|
|
|
|
use_dB_normalization: True
|
|
|
|
|
|
|
|
target_dB: -20
|
|
|
|
|
|
|
|
random_seed: 0
|
|
|
|
|
|
|
|
keep_transcription_text: False
|
|
|
|
|
|
|
|
sortagrad: True
|
|
|
|
|
|
|
|
shuffle_method: batch_shuffle
|
|
|
|
|
|
|
|
num_workers: 2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# network architecture
|
|
|
|
# network architecture
|
|
|
|
model:
|
|
|
|
model:
|
|
|
|
cmvn_file:
|
|
|
|
cmvn_file:
|
|
|
@ -76,8 +38,40 @@ model:
|
|
|
|
length_normalized_loss: false
|
|
|
|
length_normalized_loss: false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
data:
|
|
|
|
|
|
|
|
train_manifest: data/manifest.train
|
|
|
|
|
|
|
|
dev_manifest: data/manifest.dev
|
|
|
|
|
|
|
|
test_manifest: data/manifest.test-clean
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
collator:
|
|
|
|
|
|
|
|
vocab_filepath: data/lang_char/vocab.txt
|
|
|
|
|
|
|
|
unit_type: 'spm'
|
|
|
|
|
|
|
|
spm_model_prefix: 'data/lang_char/bpe_unigram_5000'
|
|
|
|
|
|
|
|
mean_std_filepath: ""
|
|
|
|
|
|
|
|
augmentation_config: conf/preprocess.yaml
|
|
|
|
|
|
|
|
batch_size: 16
|
|
|
|
|
|
|
|
raw_wav: True # use raw_wav or kaldi feature
|
|
|
|
|
|
|
|
spectrum_type: fbank #linear, mfcc, fbank
|
|
|
|
|
|
|
|
feat_dim: 80
|
|
|
|
|
|
|
|
delta_delta: False
|
|
|
|
|
|
|
|
dither: 1.0
|
|
|
|
|
|
|
|
target_sample_rate: 16000
|
|
|
|
|
|
|
|
max_freq: None
|
|
|
|
|
|
|
|
n_fft: None
|
|
|
|
|
|
|
|
stride_ms: 10.0
|
|
|
|
|
|
|
|
window_ms: 25.0
|
|
|
|
|
|
|
|
use_dB_normalization: True
|
|
|
|
|
|
|
|
target_dB: -20
|
|
|
|
|
|
|
|
random_seed: 0
|
|
|
|
|
|
|
|
keep_transcription_text: False
|
|
|
|
|
|
|
|
sortagrad: True
|
|
|
|
|
|
|
|
shuffle_method: batch_shuffle
|
|
|
|
|
|
|
|
num_workers: 2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
training:
|
|
|
|
training:
|
|
|
|
n_epoch: 120
|
|
|
|
n_epoch: 70
|
|
|
|
accum_grad: 8
|
|
|
|
accum_grad: 8
|
|
|
|
global_grad_clip: 3.0
|
|
|
|
global_grad_clip: 3.0
|
|
|
|
optim: adam
|
|
|
|
optim: adam
|
|
|
@ -98,13 +92,7 @@ decoding:
|
|
|
|
batch_size: 64
|
|
|
|
batch_size: 64
|
|
|
|
error_rate_type: wer
|
|
|
|
error_rate_type: wer
|
|
|
|
decoding_method: attention # 'attention', 'ctc_greedy_search', 'ctc_prefix_beam_search', 'attention_rescoring'
|
|
|
|
decoding_method: attention # 'attention', 'ctc_greedy_search', 'ctc_prefix_beam_search', 'attention_rescoring'
|
|
|
|
lang_model_path: data/lm/common_crawl_00.prune01111.trie.klm
|
|
|
|
|
|
|
|
alpha: 2.5
|
|
|
|
|
|
|
|
beta: 0.3
|
|
|
|
|
|
|
|
beam_size: 10
|
|
|
|
beam_size: 10
|
|
|
|
cutoff_prob: 1.0
|
|
|
|
|
|
|
|
cutoff_top_n: 0
|
|
|
|
|
|
|
|
num_proc_bsearch: 8
|
|
|
|
|
|
|
|
ctc_weight: 0.5 # ctc weight for attention rescoring decode mode.
|
|
|
|
ctc_weight: 0.5 # ctc weight for attention rescoring decode mode.
|
|
|
|
decoding_chunk_size: -1 # decoding chunk size. Defaults to -1.
|
|
|
|
decoding_chunk_size: -1 # decoding chunk size. Defaults to -1.
|
|
|
|
# <0: for decoding, use full chunk.
|
|
|
|
# <0: for decoding, use full chunk.
|
|
|
|