Update test_srt.py

Since SRT files are mostly generated to watch with the video, rather than having an output displayed on the screen ...I propose these changes.
I hope "utf-8" encoding declaration can properly display all the language fonts across all the models in the generated srt file...Please make sure before committing.

python/example/test_srt.py

@@ -52,5 +52,6 @@ def transcribe():
                    end=datetime.timedelta(seconds=line[-1]['end']))
            subs.append(s)
     return subs
-
-print (srt.compose(transcribe()))
+subtitle = open('subtitle.srt', 'w', encoding="utf-8")
+print ((srt.compose(transcribe())), file = subtitle)
+subtitle.close()