From 28a19dbc4e85d3b8a4ec2ef7483bba64d422b43f Mon Sep 17 00:00:00 2001
From: aky15 <ankeyu.aky@11.17.44.249>
Date: 星期三, 12 四月 2023 18:03:06 +0800
Subject: [PATCH] Merge remote-tracking branch 'origin/main' into dev_aky
---
funasr/bin/asr_inference_uniasr.py | 181 +++++++--------------------------------------
1 files changed, 28 insertions(+), 153 deletions(-)
diff --git a/funasr/bin/asr_inference_uniasr.py b/funasr/bin/asr_inference_uniasr.py
index 515c0d4..4aea720 100644
--- a/funasr/bin/asr_inference_uniasr.py
+++ b/funasr/bin/asr_inference_uniasr.py
@@ -37,16 +37,13 @@
from funasr.models.frontend.wav_frontend import WavFrontend
-header_colors = '\033[95m'
-end_colors = '\033[0m'
-
class Speech2Text:
"""Speech2Text class
Examples:
>>> import soundfile
- >>> speech2text = Speech2Text("asr_config.yml", "asr.pth")
+ >>> speech2text = Speech2Text("asr_config.yml", "asr.pb")
>>> audio, rate = soundfile.read("speech.wav")
>>> speech2text(audio)
[(text, token, token_int, hypothesis object), ...]
@@ -96,11 +93,14 @@
else:
decoder = asr_model.decoder2
- ctc = CTCPrefixScorer(ctc=asr_model.ctc, eos=asr_model.eos)
+ if asr_model.ctc != None:
+ ctc = CTCPrefixScorer(ctc=asr_model.ctc, eos=asr_model.eos)
+ scorers.update(
+ ctc=ctc
+ )
token_list = asr_model.token_list
scorers.update(
decoder=decoder,
- ctc=ctc,
length_bonus=LengthBonus(len(token_list)),
)
@@ -258,6 +258,7 @@
# Change integer-ids to tokens
token = self.converter.ids2tokens(token_int)
+ token = list(filter(lambda x: x != "<gbg>", token))
if self.tokenizer is not None:
text = self.tokenizer.tokens2text(token)
@@ -268,150 +269,6 @@
assert check_return_type(results)
return results
-
-# def inference(
-# maxlenratio: float,
-# minlenratio: float,
-# batch_size: int,
-# beam_size: int,
-# ngpu: int,
-# ctc_weight: float,
-# lm_weight: float,
-# penalty: float,
-# log_level: Union[int, str],
-# data_path_and_name_and_type,
-# asr_train_config: Optional[str],
-# asr_model_file: Optional[str],
-# ngram_file: Optional[str] = None,
-# cmvn_file: Optional[str] = None,
-# raw_inputs: Union[np.ndarray, torch.Tensor] = None,
-# lm_train_config: Optional[str] = None,
-# lm_file: Optional[str] = None,
-# token_type: Optional[str] = None,
-# key_file: Optional[str] = None,
-# word_lm_train_config: Optional[str] = None,
-# bpemodel: Optional[str] = None,
-# allow_variable_data_keys: bool = False,
-# streaming: bool = False,
-# output_dir: Optional[str] = None,
-# dtype: str = "float32",
-# seed: int = 0,
-# ngram_weight: float = 0.9,
-# nbest: int = 1,
-# num_workers: int = 1,
-# token_num_relax: int = 1,
-# decoding_ind: int = 0,
-# decoding_mode: str = "model1",
-# **kwargs,
-# ):
-# assert check_argument_types()
-# if batch_size > 1:
-# raise NotImplementedError("batch decoding is not implemented")
-# if word_lm_train_config is not None:
-# raise NotImplementedError("Word LM is not implemented")
-# if ngpu > 1:
-# raise NotImplementedError("only single GPU decoding is supported")
-#
-# logging.basicConfig(
-# level=log_level,
-# format="%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s",
-# )
-#
-# if ngpu >= 1 and torch.cuda.is_available():
-# device = "cuda"
-# else:
-# device = "cpu"
-#
-# # 1. Set random-seed
-# set_all_random_seed(seed)
-#
-# # 2. Build speech2text
-# speech2text_kwargs = dict(
-# asr_train_config=asr_train_config,
-# asr_model_file=asr_model_file,
-# cmvn_file=cmvn_file,
-# lm_train_config=lm_train_config,
-# lm_file=lm_file,
-# ngram_file=ngram_file,
-# token_type=token_type,
-# bpemodel=bpemodel,
-# device=device,
-# maxlenratio=maxlenratio,
-# minlenratio=minlenratio,
-# dtype=dtype,
-# beam_size=beam_size,
-# ctc_weight=ctc_weight,
-# lm_weight=lm_weight,
-# ngram_weight=ngram_weight,
-# penalty=penalty,
-# nbest=nbest,
-# streaming=streaming,
-# token_num_relax=token_num_relax,
-# decoding_ind=decoding_ind,
-# decoding_mode=decoding_mode,
-# )
-# speech2text = Speech2Text(**speech2text_kwargs)
-#
-# # 3. Build data-iterator
-# loader = ASRTask.build_streaming_iterator(
-# data_path_and_name_and_type,
-# dtype=dtype,
-# batch_size=batch_size,
-# key_file=key_file,
-# num_workers=num_workers,
-# preprocess_fn=ASRTask.build_preprocess_fn(speech2text.asr_train_args, False),
-# collate_fn=ASRTask.build_collate_fn(speech2text.asr_train_args, False),
-# allow_variable_data_keys=allow_variable_data_keys,
-# inference=True,
-# )
-#
-# finish_count = 0
-# file_count = 1
-# # 7 .Start for-loop
-# # FIXME(kamo): The output format should be discussed about
-# asr_result_list = []
-# if output_dir is not None:
-# writer = DatadirWriter(output_dir)
-# else:
-# writer = None
-#
-# for keys, batch in loader:
-# assert isinstance(batch, dict), type(batch)
-# assert all(isinstance(s, str) for s in keys), keys
-# _bs = len(next(iter(batch.values())))
-# assert len(keys) == _bs, f"{len(keys)} != {_bs}"
-# #batch = {k: v[0] for k, v in batch.items() if not k.endswith("_lengths")}
-#
-# # N-best list of (text, token, token_int, hyp_object)
-# try:
-# results = speech2text(**batch)
-# except TooShortUttError as e:
-# logging.warning(f"Utterance {keys} {e}")
-# hyp = Hypothesis(score=0.0, scores={}, states={}, yseq=[])
-# results = [[" ", ["<space>"], [2], hyp]] * nbest
-#
-# # Only supporting batch_size==1
-# key = keys[0]
-# logging.info(f"Utterance: {key}")
-# for n, (text, token, token_int, hyp) in zip(range(1, nbest + 1), results):
-# # Create a directory: outdir/{n}best_recog
-# if writer is not None:
-# ibest_writer = writer[f"{n}best_recog"]
-#
-# # Write the result to each file
-# ibest_writer["token"][key] = " ".join(token)
-# ibest_writer["token_int"][key] = " ".join(map(str, token_int))
-# ibest_writer["score"][key] = str(hyp.score)
-#
-# if text is not None:
-# text_postprocessed = postprocess_utils.sentence_postprocess(token)
-# item = {'key': key, 'value': text_postprocessed}
-# asr_result_list.append(item)
-# finish_count += 1
-# asr_utils.print_progress(finish_count / file_count)
-# if writer is not None:
-# ibest_writer["text"][key] = text
-# return asr_result_list
def inference(
maxlenratio: float,
@@ -518,6 +375,7 @@
token_num_relax: int = 1,
decoding_ind: int = 0,
decoding_mode: str = "model1",
+ param_dict: dict = None,
**kwargs,
):
assert check_argument_types()
@@ -537,6 +395,19 @@
device = "cuda"
else:
device = "cpu"
+
+ if param_dict is not None and "decoding_model" in param_dict:
+ if param_dict["decoding_model"] == "fast":
+ decoding_ind = 0
+ decoding_mode = "model1"
+ elif param_dict["decoding_model"] == "normal":
+ decoding_ind = 0
+ decoding_mode = "model2"
+ elif param_dict["decoding_model"] == "offline":
+ decoding_ind = 1
+ decoding_mode = "model2"
+ else:
+ raise NotImplementedError("unsupported decoding model {}".format(param_dict["decoding_model"]))
# 1. Set random-seed
set_all_random_seed(seed)
@@ -571,6 +442,9 @@
def _forward(data_path_and_name_and_type,
raw_inputs: Union[np.ndarray, torch.Tensor] = None,
output_dir_v2: Optional[str] = None,
+ fs: dict = None,
+ param_dict: dict = None,
+ **kwargs,
):
# 3. Build data-iterator
if data_path_and_name_and_type is None and raw_inputs is not None:
@@ -580,6 +454,7 @@
loader = ASRTask.build_streaming_iterator(
data_path_and_name_and_type,
dtype=dtype,
+ fs=fs,
batch_size=batch_size,
key_file=key_file,
num_workers=num_workers,
@@ -613,7 +488,7 @@
except TooShortUttError as e:
logging.warning(f"Utterance {keys} {e}")
hyp = Hypothesis(score=0.0, scores={}, states={}, yseq=[])
- results = [[" ", ["<space>"], [2], hyp]] * nbest
+ results = [[" ", ["sil"], [2], hyp]] * nbest
# Only supporting batch_size==1
key = keys[0]
@@ -629,13 +504,13 @@
ibest_writer["score"][key] = str(hyp.score)
if text is not None:
- text_postprocessed = postprocess_utils.sentence_postprocess(token)
+ text_postprocessed, word_lists = postprocess_utils.sentence_postprocess(token)
item = {'key': key, 'value': text_postprocessed}
asr_result_list.append(item)
finish_count += 1
asr_utils.print_progress(finish_count / file_count)
if writer is not None:
- ibest_writer["text"][key] = text
+ ibest_writer["text"][key] = " ".join(word_lists)
return asr_result_list
return _forward
--
Gitblit v1.9.1