From 410a85402db06bf36a9d7acec5dc922012951242 Mon Sep 17 00:00:00 2001
From: shixian.shi <shixian.shi@alibaba-inc.com>
Date: 星期二, 06 二月 2024 17:55:35 +0800
Subject: [PATCH] bug fix for numpy input
---
funasr/auto/auto_model.py | 12 +++++-------
1 files changed, 5 insertions(+), 7 deletions(-)
diff --git a/funasr/auto/auto_model.py b/funasr/auto/auto_model.py
index d99fc56..8e00703 100644
--- a/funasr/auto/auto_model.py
+++ b/funasr/auto/auto_model.py
@@ -88,7 +88,7 @@
class AutoModel:
def __init__(self, **kwargs):
- if kwargs.get("disable_log", False):
+ if not kwargs.get("disable_log", False):
tables.print()
model, kwargs = self.build_model(**kwargs)
@@ -121,9 +121,6 @@
if spk_mode not in ["default", "vad_segment", "punc_segment"]:
logging.error("spk_mode should be one of default, vad_segment and punc_segment.")
self.spk_mode = spk_mode
- self.preset_spk_num = kwargs.get("preset_spk_num", None)
- if self.preset_spk_num:
- logging.warning("Using preset speaker number: {}".format(self.preset_spk_num))
self.kwargs = kwargs
self.model = model
@@ -388,11 +385,11 @@
result["text"] = punc_res[0]["text"]
# speaker embedding cluster after resorted
- if self.spk_model is not None:
+ if self.spk_model is not None and kwargs.get('return_spk_res', True):
all_segments = sorted(all_segments, key=lambda x: x[0])
spk_embedding = result['spk_embedding']
- labels = self.cb_model(spk_embedding.cpu(), oracle_num=self.preset_spk_num)
- del result['spk_embedding']
+ labels = self.cb_model(spk_embedding.cpu(), oracle_num=kwargs.get('preset_spk_num', None))
+ # del result['spk_embedding']
sv_output = postprocess(all_segments, None, labels, spk_embedding.cpu())
if self.spk_mode == 'vad_segment': # recover sentence_list
sentence_list = []
@@ -412,6 +409,7 @@
result['timestamp'], \
result['raw_text'])
result['sentence_info'] = sentence_list
+ del result['spk_embedding']
result["key"] = key
results_ret_list.append(result)
--
Gitblit v1.9.1