Move STT timing to start at recording end

neon_speech/service.py

@@ -34,6 +34,8 @@
 from tempfile import mkstemp
 from threading import Lock, Event
 from time import time
+
+from ovos_utils.process_utils import ProcessState
 from pydub import AudioSegment
 from speech_recognition import AudioData
 from neon_utils.file_utils import decode_base64_string_to_file
@@ -135,9 +137,9 @@ def __init__(self, ready_hook=on_ready, error_hook=on_error,
             LOG.info("Skipping api_stt init")
             self.api_stt = None
 
-    def _record_begin(self):
+    def _record_end_signal(self):
         self._stt_stopwatch.start()
-        OVOSDinkumVoiceService._record_begin(self)
+        OVOSDinkumVoiceService._record_end_signal(self)
 
     def _stt_text(self, text: str, stt_context: dict):
         self._stt_stopwatch.stop()