From 7b4f2c4574ca0c2980d93e1790f21e2fcaf6fd43 Mon Sep 17 00:00:00 2001
From: haoneng.lhn <haoneng.lhn@alibaba-inc.com>
Date: 星期一, 26 六月 2023 17:51:48 +0800
Subject: [PATCH] update soundfile version for loading mp3 file
---
setup.py | 15 ++++++++++-----
1 files changed, 10 insertions(+), 5 deletions(-)
diff --git a/setup.py b/setup.py
index fdbfad9..994b088 100644
--- a/setup.py
+++ b/setup.py
@@ -17,10 +17,10 @@
"humanfriendly",
"scipy>=1.4.1",
# "filelock",
- "librosa==0.8.1",
+ "librosa",
"jamo==0.4.1", # For kss
"PyYAML>=5.1.2",
- "soundfile>=0.10.2",
+ "soundfile>=0.11.0",
"h5py>=2.10.0",
"kaldiio>=2.17.0",
"torch_complex",
@@ -40,7 +40,12 @@
"g2p",
# PAI
"oss2",
- "kaldi-native-fbank",
+ # "kaldi-native-fbank",
+ # timestamp
+ "edit-distance",
+ # textgrid
+ "textgrid",
+ "protobuf==3.20.0",
],
# train: The modules invoked when training only.
"train": [
@@ -76,7 +81,7 @@
# "gtn==0.0.0",
],
"setup": [
- "numpy<=1.21.3",
+ "numpy",
"pytest-runner",
],
"test": [
@@ -121,7 +126,7 @@
name="funasr",
version=version,
url="https://github.com/alibaba-damo-academy/FunASR.git",
- author="Speech Lab, Alibaba Group, China",
+ author="Speech Lab of DAMO Academy, Alibaba Group",
author_email="funasr@list.alibaba-inc.com",
description="FunASR: A Fundamental End-to-End Speech Recognition Toolkit",
long_description=open(os.path.join(dirname, "README.md"), encoding="utf-8").read(),
--
Gitblit v1.9.1