diff --git a/yt_dlp/extractor/abematv.py b/yt_dlp/extractor/abematv.py
index 7552e3e57..f611c1f2c 100644
--- a/yt_dlp/extractor/abematv.py
+++ b/yt_dlp/extractor/abematv.py
@@ -156,7 +156,7 @@ class AbemaTVBaseIE(InfoExtractor):
     def _generate_aks(cls, deviceid):
         deviceid = deviceid.encode('utf-8')
         # add 1 hour and then drop minute and secs
-        ts_1hour = int((time_seconds(hours=9) // 3600 + 1) * 3600)
+        ts_1hour = int((time_seconds() // 3600 + 1) * 3600)
         time_struct = time.gmtime(ts_1hour)
         ts_1hour_str = str(ts_1hour).encode('utf-8')
 
@@ -190,6 +190,16 @@ class AbemaTVBaseIE(InfoExtractor):
         if self._USERTOKEN:
             return self._USERTOKEN
 
+        username, _ = self._get_login_info()
+        AbemaTVBaseIE._USERTOKEN = username and self.cache.load(self._NETRC_MACHINE, username)
+        if AbemaTVBaseIE._USERTOKEN:
+            # try authentication with locally stored token
+            try:
+                self._get_media_token(True)
+                return
+            except ExtractorError as e:
+                self.report_warning(f'Failed to login with cached user token; obtaining a fresh one ({e})')
+
         AbemaTVBaseIE._DEVICE_ID = str(uuid.uuid4())
         aks = self._generate_aks(self._DEVICE_ID)
         user_data = self._download_json(
@@ -300,6 +310,11 @@ class AbemaTVIE(AbemaTVBaseIE):
     _TIMETABLE = None
 
     def _perform_login(self, username, password):
+        self._get_device_token()
+        if self.cache.load(self._NETRC_MACHINE, username) and self._get_media_token():
+            self.write_debug('Skipping logging in')
+            return
+
         if '@' in username:  # don't strictly check if it's email address or not
             ep, method = 'user/email', 'email'
         else:
@@ -319,6 +334,7 @@ class AbemaTVIE(AbemaTVBaseIE):
 
         AbemaTVBaseIE._USERTOKEN = login_response['token']
         self._get_media_token(True)
+        self.cache.store(self._NETRC_MACHINE, username, AbemaTVBaseIE._USERTOKEN)
 
     def _real_extract(self, url):
         # starting download using infojson from this extractor is undefined behavior,