From 4f98546f3693482f8f34aa5f11ced31381c58724 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期五, 05 一月 2024 16:55:07 +0800
Subject: [PATCH] load_audio_text_image_video
---
funasr/models/paraformer_streaming/model.py | 6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/funasr/models/paraformer_streaming/model.py b/funasr/models/paraformer_streaming/model.py
index a57c927..e57bc34 100644
--- a/funasr/models/paraformer_streaming/model.py
+++ b/funasr/models/paraformer_streaming/model.py
@@ -40,7 +40,7 @@
@contextmanager
def autocast(enabled=True):
yield
-from funasr.utils.load_utils import load_audio_and_text_image_video, extract_fbank
+from funasr.utils.load_utils import load_audio_text_image_video, extract_fbank
from funasr.utils import postprocess_utils
from funasr.utils.datadir_writer import DatadirWriter
from funasr.utils.timestamp_tools import ts_prediction_lfr6_standard
@@ -483,7 +483,7 @@
meta_data = {}
# extract fbank feats
time1 = time.perf_counter()
- audio_sample_list = load_audio_and_text_image_video(data_in, fs=self.frontend.fs, audio_fs=kwargs.get("fs", 16000))
+ audio_sample_list = load_audio_text_image_video(data_in, fs=self.frontend.fs, audio_fs=kwargs.get("fs", 16000))
time2 = time.perf_counter()
meta_data["load_data"] = f"{time2 - time1:0.3f}"
speech, speech_lengths = extract_fbank(audio_sample_list, data_type=kwargs.get("data_type", "sound"), frontend=self.frontend)
@@ -761,7 +761,7 @@
meta_data = {}
# extract fbank feats
time1 = time.perf_counter()
- audio_sample_list = load_audio_and_text_image_video(data_in, fs=self.frontend.fs, audio_fs=kwargs.get("fs", 16000))
+ audio_sample_list = load_audio_text_image_video(data_in, fs=self.frontend.fs, audio_fs=kwargs.get("fs", 16000))
time2 = time.perf_counter()
meta_data["load_data"] = f"{time2 - time1:0.3f}"
speech, speech_lengths = extract_fbank(audio_sample_list, data_type=kwargs.get("data_type", "sound"),
--
Gitblit v1.9.1