From fce4e1d1b48f23cd8332e60afce3df8d6209a6a7 Mon Sep 17 00:00:00 2001
From: gaochangfeng <54253717+gaochangfeng@users.noreply.github.com>
Date: 星期四, 11 四月 2024 14:59:22 +0800
Subject: [PATCH] SenseVoice对富文本解码的参数 (#1608)
---
funasr/bin/compute_audio_cmvn.py | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/funasr/bin/compute_audio_cmvn.py b/funasr/bin/compute_audio_cmvn.py
index 6282e70..cd64329 100644
--- a/funasr/bin/compute_audio_cmvn.py
+++ b/funasr/bin/compute_audio_cmvn.py
@@ -59,6 +59,7 @@
dataset_conf = kwargs.get("dataset_conf")
dataset_conf["batch_type"] = "example"
dataset_conf["batch_size"] = 1
+ dataset_conf["num_workers"] = os.cpu_count() or 32
batch_sampler_train = batch_sampler_class(dataset_train, is_training=False, **dataset_conf)
dataloader_train = torch.utils.data.DataLoader(dataset_train, collate_fn=dataset_train.collator, **batch_sampler_train)
--
Gitblit v1.9.1