From 9fcb3cc06b4e324f0913d2f61b89becc2baeef1b Mon Sep 17 00:00:00 2001
From: hnluo <haoneng.lhn@alibaba-inc.com>
Date: 星期一, 11 九月 2023 17:40:03 +0800
Subject: [PATCH] Merge pull request #932 from alibaba-damo-academy/dev_lhn
---
funasr/bin/asr_inference_launch.py | 33 +++++++++++++++++++++++----------
1 files changed, 23 insertions(+), 10 deletions(-)
diff --git a/funasr/bin/asr_inference_launch.py b/funasr/bin/asr_inference_launch.py
index fd0eecd..6fa57a7 100644
--- a/funasr/bin/asr_inference_launch.py
+++ b/funasr/bin/asr_inference_launch.py
@@ -236,6 +236,7 @@
timestamp_infer_config: Union[Path, str] = None,
timestamp_model_file: Union[Path, str] = None,
param_dict: dict = None,
+ decoding_ind: int = 0,
**kwargs,
):
ncpu = kwargs.get("ncpu", 1)
@@ -260,8 +261,6 @@
hotword_list_or_file = None
clas_scale = 1.0
- if kwargs.get("device", None) == "cpu":
- ngpu = 0
if ngpu >= 1 and torch.cuda.is_available():
device = "cuda"
else:
@@ -292,6 +291,7 @@
nbest=nbest,
hotword_list_or_file=hotword_list_or_file,
clas_scale=clas_scale,
+ decoding_ind=decoding_ind,
)
speech2text = Speech2TextParaformer(**speech2text_kwargs)
@@ -314,6 +314,7 @@
**kwargs,
):
+ decoding_ind = None
hotword_list_or_file = None
if param_dict is not None:
hotword_list_or_file = param_dict.get('hotword')
@@ -321,6 +322,8 @@
hotword_list_or_file = kwargs['hotword']
if hotword_list_or_file is not None or 'hotword' in kwargs:
speech2text.hotword_list = speech2text.generate_hotwords_list(hotword_list_or_file)
+ if param_dict is not None and "decoding_ind" in param_dict:
+ decoding_ind = param_dict["decoding_ind"]
# 3. Build data-iterator
if data_path_and_name_and_type is None and raw_inputs is not None:
@@ -367,6 +370,7 @@
# N-best list of (text, token, token_int, hyp_object)
time_beg = time.time()
+ batch["decoding_ind"] = decoding_ind
results = speech2text(**batch)
if len(results) < 1:
hyp = Hypothesis(score=0.0, scores={}, states={}, yseq=[])
@@ -565,6 +569,8 @@
if 'hotword' in kwargs:
hotword_list_or_file = kwargs['hotword']
+ speech2vadsegment.vad_model.vad_opts.max_single_segment_time = kwargs.get("max_single_segment_time", 60000)
+ batch_size_token_threshold_s = kwargs.get("batch_size_token_threshold_s", int(speech2vadsegment.vad_model.vad_opts.max_single_segment_time*0.67/1000)) * 1000
batch_size_token = kwargs.get("batch_size_token", 6000)
print("batch_size_token: ", batch_size_token)
@@ -647,8 +653,7 @@
beg_idx = 0
for j, _ in enumerate(range(0, n)):
batch_size_token_ms_cum += (sorted_data[j][0][1] - sorted_data[j][0][0])
- if j < n - 1 and (batch_size_token_ms_cum + sorted_data[j + 1][0][1] - sorted_data[j + 1][0][
- 0]) < batch_size_token_ms:
+ if j < n - 1 and (batch_size_token_ms_cum + sorted_data[j + 1][0][1] - sorted_data[j + 1][0][0]) < batch_size_token_ms and (sorted_data[j + 1][0][1] - sorted_data[j + 1][0][0]) < batch_size_token_threshold_s:
continue
batch_size_token_ms_cum = 0
end_idx = j + 1
@@ -1291,6 +1296,7 @@
quantize_dtype: Optional[str] = "float16",
streaming: Optional[bool] = False,
simu_streaming: Optional[bool] = False,
+ full_utt: Optional[bool] = False,
chunk_size: Optional[int] = 16,
left_context: Optional[int] = 16,
right_context: Optional[int] = 0,
@@ -1340,7 +1346,7 @@
format="%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s",
)
- if ngpu >= 1:
+ if ngpu >= 1 and torch.cuda.is_available():
device = "cuda"
else:
device = "cpu"
@@ -1367,14 +1373,12 @@
quantize_dtype=quantize_dtype,
streaming=streaming,
simu_streaming=simu_streaming,
+ full_utt=full_utt,
chunk_size=chunk_size,
left_context=left_context,
right_context=right_context,
)
- speech2text = Speech2TextTransducer.from_pretrained(
- model_tag=model_tag,
- **speech2text_kwargs,
- )
+ speech2text = Speech2TextTransducer(**speech2text_kwargs)
def _forward(data_path_and_name_and_type,
raw_inputs: Union[np.ndarray, torch.Tensor] = None,
@@ -1420,7 +1424,7 @@
_end = (i + 1) * speech2text._ctx
speech2text.streaming_decode(
- speech[i * speech2text._ctx: _end], is_final=False
+ speech[i * speech2text._ctx: _end + speech2text._right_ctx], is_final=False
)
final_hyps = speech2text.streaming_decode(
@@ -1428,6 +1432,8 @@
)
elif speech2text.simu_streaming:
final_hyps = speech2text.simu_streaming_decode(**batch)
+ elif speech2text.full_utt:
+ final_hyps = speech2text.full_utt_decode(**batch)
else:
final_hyps = speech2text(**batch)
@@ -1786,6 +1792,12 @@
default=1,
help="The batch size for inference",
)
+ group.add_argument(
+ "--decoding_ind",
+ type=int,
+ default=0,
+ help="chunk select for chunk encoder",
+ )
group.add_argument("--nbest", type=int, default=5, help="Output N-best hypotheses")
group.add_argument("--beam_size", type=int, default=20, help="Beam size")
group.add_argument("--penalty", type=float, default=0.0, help="Insertion penalty")
@@ -1816,6 +1828,7 @@
group.add_argument("--ngram_weight", type=float, default=0.9, help="ngram weight")
group.add_argument("--streaming", type=str2bool, default=False)
group.add_argument("--simu_streaming", type=str2bool, default=False)
+ group.add_argument("--full_utt", type=str2bool, default=False)
group.add_argument("--chunk_size", type=int, default=16)
group.add_argument("--left_context", type=int, default=16)
group.add_argument("--right_context", type=int, default=0)
--
Gitblit v1.9.1