From f43da18b5e097515c12438c70fc7918c29e5d0c8 Mon Sep 17 00:00:00 2001
From: lingji-yidong <75744976+lingji-yidong@users.noreply.github.com>
Date: Mon, 19 Aug 2024 13:36:59 +0800
Subject: [PATCH] fix start timestamp in sentence_info (#2024)

Previously, the start timestamp was defined as the last character's timestamp[0] of the previous sentence. It has now been changed to the first character's timestamp[0] of the current sentence.
---
 funasr/utils/timestamp_tools.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/funasr/utils/timestamp_tools.py b/funasr/utils/timestamp_tools.py
index 3d8fd1062..369dc1bc3 100644
--- a/funasr/utils/timestamp_tools.py
+++ b/funasr/utils/timestamp_tools.py
@@ -141,6 +141,8 @@ def timestamp_sentence(
     )
     for punc_stamp_text in punc_stamp_text_list:
         punc_id, timestamp, text = punc_stamp_text
+        if sentence_start is None and timestamp is not None:
+            sentence_start = timestamp[0]
         # sentence_text += text if text is not None else ''
         if text is not None:
             if "a" <= text[0] <= "z" or "A" <= text[0] <= "Z":
@@ -183,7 +185,7 @@ def timestamp_sentence(
             sentence_text = ""
             sentence_text_seg = ""
             ts_list = []
-            sentence_start = timestamp[0] if timestamp is not None else sentence_start
+            sentence_start = None
     return res
 
 
@@ -272,4 +274,4 @@ def timestamp_sentence_en(
             sentence_text = ""
             sentence_text_seg = ""
             ts_list = []
-    return res
\ No newline at end of file
+    return res