From 8b0f765b4781b5e8c56a357cfa28fffa49e8d731 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期六, 27 七月 2024 00:44:55 +0800
Subject: [PATCH] v1.1.4
---
funasr/auto/auto_model.py | 14 ++++++++------
1 files changed, 8 insertions(+), 6 deletions(-)
diff --git a/funasr/auto/auto_model.py b/funasr/auto/auto_model.py
index a82f6ed..ca1f202 100644
--- a/funasr/auto/auto_model.py
+++ b/funasr/auto/auto_model.py
@@ -114,7 +114,7 @@
try:
from funasr.utils.version_checker import check_for_update
- check_for_update()
+ check_for_update(disable=kwargs.get("disable_update", False))
except:
pass
@@ -264,6 +264,8 @@
def inference(self, input, input_len=None, model=None, kwargs=None, key=None, **cfg):
kwargs = self.kwargs if kwargs is None else kwargs
+ if "cache" in kwargs:
+ kwargs.pop("cache")
deep_update(kwargs, cfg)
model = self.model if model is None else model
model.eval()
@@ -337,7 +339,7 @@
end_vad = time.time()
# FIX(gcf): concat the vad clips for sense vocie model for better aed
- if kwargs.get("merge_vad", False):
+ if cfg.get("merge_vad", False):
for i in range(len(res)):
res[i]["value"] = merge_vad(
res[i]["value"], kwargs.get("merge_length_s", 15) * 1000
@@ -511,8 +513,8 @@
sv_output = postprocess(all_segments, None, labels, spk_embedding.cpu())
if self.spk_mode == "vad_segment": # recover sentence_list
sentence_list = []
- for res, vadsegment in zip(restored_data, vadsegments):
- if "timestamp" not in res:
+ for rest, vadsegment in zip(restored_data, vadsegments):
+ if "timestamp" not in rest:
logging.error(
"Only 'iic/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch' \
and 'iic/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch'\
@@ -522,8 +524,8 @@
{
"start": vadsegment[0],
"end": vadsegment[1],
- "sentence": res["text"],
- "timestamp": res["timestamp"],
+ "sentence": rest["text"],
+ "timestamp": rest["timestamp"],
}
)
elif self.spk_mode == "punc_segment":
--
Gitblit v1.9.1