Fix SRT sequencing for real time, use ms for remainTime/delay for all…

scenarios/cpp/windows/captioning/captioning/captioning.cpp

@@ -56,7 +56,7 @@ class Captioning
     std::shared_ptr<AudioStreamFormat> m_format = NULL;
     std::shared_ptr<BinaryFileReader> m_callback = NULL;
     std::shared_ptr<PullAudioInputStream> m_stream = NULL;
-    int m_srtSequenceNumber = 0;
+    int m_srtSequenceNumber = 1;
     std::optional<Caption> m_previousCaption = std::nullopt;
     std::optional<Timestamp> m_previousEndTime = std::nullopt;
     bool m_previousResultIsRecognized = false;
@@ -93,7 +93,7 @@ class Captioning
         std::string retval;
         if (m_userConfig->useSubRipTextCaptionFormat)
         {
-            retval += caption.sequence + "\n";
+            retval += std::to_string(caption.sequence) + "\n";
         }
         retval += GetTimestamp(caption.begin, caption.end) + "\n";
         retval += caption.text + "\n\n";
@@ -454,10 +454,10 @@ int main(int argc, char* argv[])
 "                                     Minimum is 20. Default is 37 (30 for Chinese).\n"
 "    --lines LINES                    Set the number of lines for a caption to LINES.\n"
 "                                     Minimum is 1. Default is 2.\n"
-"    --delay SECONDS                  How many SECONDS to delay the appearance of each caption.\n"
-"                                     Minimum is 0.0. Default is 1.0.\n"
-"    --remainTime SECONDS             How many SECONDS a caption should remain on screen if it is not replaced by another.\n"
-"                                     Minimum is 0.0. Default is 1.0.\n\n"
+"    --delay MILLISECONDS             How many MILLISECONDS to delay the appearance of each caption.\n"
+"                                     Minimum is 0. Default is 1000.\n"
+"    --remainTime MILLISECONDS        How many MILLISECONDS a caption should remain on screen if it is not replaced by another.\n"
+"                                     Minimum is 0. Default is 1000.\n\n"
 "    --quiet                          Suppress console output, except errors.\n"
 "    --profanity OPTION               Valid values: raw, remove, mask\n"
 "    --threshold NUMBER               Set stable partial result threshold.\n"

scenarios/cpp/windows/captioning/captioning/captioning.vcxproj

@@ -151,7 +151,11 @@
     <None Include="packages.config" />
   </ItemGroup>
   <ItemGroup>
-    <ClInclude Include="helper.h" />
+    <ClInclude Include="binary_file_reader.h" />
+    <ClInclude Include="caption_helper.h" />
+    <ClInclude Include="string_helper.h" />
+    <ClInclude Include="user_config.h" />
+    <ClInclude Include="wav_file_reader.h" />
   </ItemGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
   <ImportGroup Label="ExtensionTargets">

scenarios/csharp/dotnetcore/captioning/captioning/Program.cs

@@ -26,7 +26,7 @@ namespace Captioning
     class Program
     {
         private UserConfig? _userConfig;
-        private int _srtSequenceNumber = 0;
+        private int _srtSequenceNumber = 1;
         private Caption? _previousCaption;
         private TimeSpan? _previousEndTime;
         private bool _previousResultIsRecognized = false;
@@ -434,10 +434,10 @@ Default output mode is offline.
                                      Minimum is 20. Default is 37 (30 for Chinese).
     --lines LINES                    Set the number of lines for a caption to LINES.
                                      Minimum is 1. Default is 2.
-    --delay SECONDS                  How many SECONDS to delay the appearance of each caption.
-                                     Minimum is 0.0. Default is 1.0.
-    --remainTime SECONDS             How many SECONDS a caption should remain on screen if it is not replaced by another.
-                                     Minimum is 0.0. Default is 1.0.
+    --delay MILLISECONDS             How many MILLISECONDS to delay the appearance of each caption.
+                                     Minimum is 0. Default is 1000.
+    --remainTime MILLISECONDS        How many MILLISECONDS a caption should remain on screen if it is not replaced by another.
+                                     Minimum is 0. Default is 1000.
     --quiet                          Suppress console output, except errors.
     --profanity OPTION               Valid values: raw, remove, mask
                                      Default is mask.

scenarios/csharp/dotnetcore/captioning/captioning/UserConfig.cs

@@ -45,9 +45,9 @@ public class UserConfig
         readonly public bool suppressConsoleOutput = false;
         /// The captioning mode. Default is offline.
         readonly public CaptioningMode captioningMode = CaptioningMode.Offline;
-        /// How long (in seconds) a caption should remain on screen. Default is 1.0.
+        /// How long (in milliseconds) a caption should remain on screen. Default is 1000.
         readonly public TimeSpan remainTime;
-        /// How long (in seconds) to delay all caption timestamps. Default is 1.0.
+        /// How long (in milliseconds) to delay all caption timestamps. Default is 1000.
         readonly public TimeSpan delay;
         /// Output captions in SubRip Text format (default is WebVTT format).
         readonly public bool useSubRipTextCaptionFormat = false;
@@ -203,27 +203,27 @@ public static UserConfig UserConfigFromArgs(string[] args, string usage)
             CaptioningMode captioningMode = CmdOptionExists(args, "--realTime") && !CmdOptionExists(args, "--offline") ? CaptioningMode.RealTime : CaptioningMode.Offline;
 
             string? strRemainTime = GetCmdOption(args, "--remainTime");
-            TimeSpan timeSpanRemainTime = TimeSpan.FromSeconds(1.0);
+            TimeSpan timeSpanRemainTime = TimeSpan.FromMilliseconds(1000);
             if (null != strRemainTime)
             {
-                double dblRemainTime = Double.Parse(strRemainTime);
-                if (dblRemainTime < 0.0)
+                int intRemainTime = Int32.Parse(strRemainTime);
+                if (intRemainTime < 0)
                 {
-                    dblRemainTime = 1.0;
+                    intRemainTime = 1000;
                 }
-                timeSpanRemainTime = TimeSpan.FromSeconds(dblRemainTime);
+                timeSpanRemainTime = TimeSpan.FromMilliseconds(intRemainTime);
             }
 
             string? strDelay = GetCmdOption(args, "--delay");
-            TimeSpan timeSpanDelay = TimeSpan.FromSeconds(1.0);
+            TimeSpan timeSpanDelay = TimeSpan.FromMilliseconds(1000);
             if (null != strDelay)
             {
-                double dblDelay = Double.Parse(strDelay);
-                if (dblDelay < 0.0)
+                int intDelay = Int32.Parse(strDelay);
+                if (intDelay < 0)
                 {
-                    dblDelay = 1.0;
+                    intDelay = 1000;
                 }
-                timeSpanDelay = TimeSpan.FromSeconds(dblDelay);
+                timeSpanDelay = TimeSpan.FromMilliseconds(intDelay);
             }
 
             string? strMaxLineLength = GetCmdOption(args, "--maxLineLength");
@@ -270,4 +270,4 @@ public static UserConfig UserConfigFromArgs(string[] args, string usage)
             );
         }
     }
-}
+}

scenarios/java/jre/console/captioning/Captioning.java

@@ -31,7 +31,7 @@
 public class Captioning
 {
     UserConfig _userConfig;
-    int _srtSequenceNumber = 0;
+    int _srtSequenceNumber = 1;
     Optional<Caption> _previousCaption = Optional.empty();
     Optional<Instant> _previousEndTime = Optional.empty();
     boolean _previousResultIsRecognized = false;

scenarios/python/console/captioning/captioning.py

@@ -65,10 +65,10 @@
                                      Minimum is 20. Default is 37 (30 for Chinese).
     --lines LINES                    Set the number of lines for a caption to LINES.
                                      Minimum is 1. Default is 2.
-    --delay SECONDS                  How many SECONDS to delay the appearance of each caption.
-                                     Minimum is 0.0. Default is 1.0.
-    --remainTime SECONDS             How many SECONDS a caption should remain on screen if it is not replaced by another.
-                                     Minimum is 0.0. Default is 1.0.
+    --delay MILLISECONDS             How many MILLISECONDS to delay the appearance of each caption.
+                                     Minimum is 0. Default is 1000.
+    --remainTime MILLISECONDS        How many MILLISECONDS a caption should remain on screen if it is not replaced by another.
+                                     Minimum is 0. Default is 1000.
     --quiet                          Suppress console output, except errors.
     --profanity OPTION               Valid values: raw, remove, mask
                                      Default is mask.
@@ -79,7 +79,7 @@
 class Captioning(object) :
     def __init__(self) :
         self._user_config = user_config_helper.user_config_from_args(USAGE)
-        self._srt_sequence_number = 0
+        self._srt_sequence_number = 1
         self._previous_caption : Optional[caption_helper.Caption] = None
         self._previous_end_time : Optional[time] = None
         self._previous_result_is_recognized = False

scenarios/python/console/captioning/user_config_helper.py

@@ -89,21 +89,21 @@ def user_config_from_args(usage : str) -> helper.Read_Only_Dict :
 
     captioning_mode = CaptioningMode.REALTIME if cmd_option_exists("--realtime") and not cmd_option_exists("--offline") else CaptioningMode.OFFLINE
 
-    td_remain_time = timedelta(seconds=1.0)
+    td_remain_time = timedelta(milliseconds=1000)
     s_remain_time = get_cmd_option("--remainTime")
     if s_remain_time is not None :
-        flt_remain_time = float(s_remain_time)
-        if flt_remain_time < 0.0 :
-            flt_remain_time = 1.0
-        td_remain_time = timedelta(seconds=flt_remain_time)
+        int_remain_time = float(s_remain_time)
+        if int_remain_time < 0 :
+            int_remain_time = 1000
+        td_remain_time = timedelta(milliseconds=int_remain_time)
 
-    td_delay = timedelta(seconds=1.0)
+    td_delay = timedelta(milliseconds=1000)
     s_delay = get_cmd_option("--delay")
     if s_delay is not None :
-        flt_delay = float(s_delay)
-        if flt_delay < 0.0 :
-            flt_delay = 1.0
-        td_delay = timedelta(seconds=flt_delay)
+        int_delay = float(s_delay)
+        if int_delay < 0 :
+            int_delay = 1000
+        td_delay = timedelta(milliseconds=int_delay)
 
     int_max_line_length = helper.DEFAULT_MAX_LINE_LENGTH_SBCS
     s_max_line_length = get_cmd_option("--maxLineLength")