From 3c349ac0531b07239f37b81254f8568ab80e3f6a Mon Sep 17 00:00:00 2001
From: Han Zhang <45134013+holazzer@users.noreply.github.com>
Date: 星期二, 18 三月 2025 11:45:37 +0800
Subject: [PATCH] fix: use converted token_ids for alignment for sensevoice model with timestamp output (#2429)
---
funasr/datasets/llm_datasets/datasets.py | 2 --
1 files changed, 0 insertions(+), 2 deletions(-)
diff --git a/funasr/datasets/llm_datasets/datasets.py b/funasr/datasets/llm_datasets/datasets.py
index b660554..61caded 100644
--- a/funasr/datasets/llm_datasets/datasets.py
+++ b/funasr/datasets/llm_datasets/datasets.py
@@ -64,8 +64,6 @@
def __getitem__(self, index):
item = self.index_ds[index]
- # import pdb;
- # pdb.set_trace()
source = item["source"]
data_src = load_audio_text_image_video(source, fs=self.fs)
if self.preprocessor_speech:
--
Gitblit v1.9.1