[ASR]cherrypick change optimizer and fix import error, test=asr (#3049)

* optional tokenizer and fix some doc

* cherry pick with pr:3040
pull/3057/head r1.4.0
zxcd 2 years ago committed by GitHub
parent 4d1787dcf7
commit d103cb8f9e
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -4,6 +4,7 @@
* paddle version: develop (commit id: daea892c67e85da91906864de40ce9f6f1b893ae) * paddle version: develop (commit id: daea892c67e85da91906864de40ce9f6f1b893ae)
* paddlespeech version: develop (commit id: c14b4238b256693281e59605abff7c9435b3e2b2) * paddlespeech version: develop (commit id: c14b4238b256693281e59605abff7c9435b3e2b2)
* paddlenlp version: 2.5.2
## Device ## Device
* python: 3.7 * python: 3.7

@ -83,7 +83,7 @@ dnn_neurons: 1024
freeze_wav2vec: False freeze_wav2vec: False
dropout: 0.15 dropout: 0.15
tokenizer: !apply:transformers.BertTokenizer.from_pretrained tokenizer: !apply:paddlenlp.transformers.AutoTokenizer.from_pretrained
pretrained_model_name_or_path: bert-base-chinese pretrained_model_name_or_path: bert-base-chinese
# bert-base-chinese tokens length # bert-base-chinese tokens length
output_neurons: 21128 output_neurons: 21128

@ -21,7 +21,7 @@ import glob
import logging import logging
import os import os
from paddlespeech.s2t.models.wav2vec2.io.dataio import read_audio from paddlespeech.s2t.io.speechbrain.dataio import read_audio
logger = logging.getLogger(__name__) logger = logging.getLogger(__name__)

@ -1,7 +1,7 @@
#!/bin/bash #!/bin/bash
stage=-1 stage=-1
stop_stage=-1 stop_stage=3
dict_dir=data/lang_char dict_dir=data/lang_char
. ${MAIN_ROOT}/utils/parse_options.sh || exit -1; . ${MAIN_ROOT}/utils/parse_options.sh || exit -1;

Loading…
Cancel
Save