From b57b98364ff60ae0119b2e8d92471316bb4e504f Mon Sep 17 00:00:00 2001
From: zhifu gao <zhifu.gzf@alibaba-inc.com>
Date: 星期三, 22 十一月 2023 00:36:35 +0800
Subject: [PATCH] funasr v2 setup (#1106)
---
funasr/datasets/large_datasets/dataset.py | 6 ++++--
1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/funasr/datasets/large_datasets/dataset.py b/funasr/datasets/large_datasets/dataset.py
index adfe4f6..d3489c1 100644
--- a/funasr/datasets/large_datasets/dataset.py
+++ b/funasr/datasets/large_datasets/dataset.py
@@ -7,7 +7,8 @@
import torch.distributed as dist
import torchaudio
import numpy as np
-import soundfile
+# import librosa
+import librosa
from kaldiio import ReadHelper
from torch.utils.data import IterableDataset
@@ -128,7 +129,8 @@
try:
waveform, sampling_rate = torchaudio.load(path)
except:
- waveform, sampling_rate = soundfile.read(path, dtype='float32')
+ # waveform, sampling_rate = librosa.load(path, dtype='float32')
+ waveform, sampling_rate = librosa.load(path, dtype='float32')
if waveform.ndim == 2:
waveform = waveform[:, 0]
waveform = np.expand_dims(waveform, axis=0)
--
Gitblit v1.9.1