Add Opus DRED and OSCE AI audio effects

Author: hwangsihu

ChangeLog.txt

@@ -4,11 +4,17 @@
 
 Version 5.22.1, unreleased
 Default Qt Client
--
+- Option to enable AI packet loss recovery (DRED), making outgoing voice more resilient on unstable networks
+- Option to enable AI speech enhancement (OSCE) to improve clarity of received voice
 Android Client
--
+- Option to enable AI packet loss recovery (DRED), making outgoing voice more resilient on unstable networks
+- Option to enable AI speech enhancement (OSCE) to improve clarity of received voice
 iOS Client
--
+- Option to enable AI packet loss recovery (DRED), making outgoing voice more resilient on unstable networks
+- Option to enable AI speech enhancement (OSCE) to improve clarity of received voice
+Accessible Windows Client
+- Option to enable AI packet loss recovery (DRED), making outgoing voice more resilient on unstable networks
+- Option to enable AI speech enhancement (OSCE) to improve clarity of received voice
 Server
 -
 

Client/TeamTalkAndroid/src/main/java/dk/bearware/backend/TeamTalkService.java

@@ -75,6 +75,7 @@
 import javax.xml.xpath.XPathExpressionException;
 import javax.xml.xpath.XPathFactory;
 
+import dk.bearware.AIAudioEffect;
 import dk.bearware.AudioPreprocessor;
 import dk.bearware.AudioPreprocessorType;
 import dk.bearware.Channel;
@@ -869,6 +870,12 @@ private void setupAudioPreprocessor() {
             int gain = prefs.getInt(Preferences.PREF_SOUNDSYSTEM_MICROPHONEGAIN, SoundLevel.SOUND_GAIN_DEFAULT);
             ttclient.setSoundInputGainLevel(gain);
         }
+
+        SharedPreferences prefs = PreferenceManager.getDefaultSharedPreferences(getApplicationContext());
+        AIAudioEffect aiEffect = new AIAudioEffect();
+        aiEffect.bEnableDRED = prefs.getBoolean(Preferences.PREF_SOUNDSYSTEM_AI_DRED, false);
+        aiEffect.bEnableOSCE = prefs.getBoolean(Preferences.PREF_SOUNDSYSTEM_AI_OSCE, false);
+        ttclient.setAIAudioEffect(aiEffect);
     }
 
     @Override

Client/TeamTalkAndroid/src/main/java/dk/bearware/data/Preferences.java

@@ -44,7 +44,9 @@ public class Preferences {
             PREF_SOUNDSYSTEM_MUTE_ON_TRANSMISSION = "mute_speakers_on_tx_checkbox",
             PREF_SOUNDSYSTEM_SPEAKERPHONE = "speakerphone_checkbox",
             PREF_SOUNDSYSTEM_BLUETOOTH_HEADSET = "bluetooth_headset_checkbox",
-            PREF_SOUNDSYSTEM_VOICEPROCESSING = "voiceprocessing_checkbox";
+            PREF_SOUNDSYSTEM_VOICEPROCESSING = "voiceprocessing_checkbox",
+            PREF_SOUNDSYSTEM_AI_DRED = "ai_dred_checkbox",
+            PREF_SOUNDSYSTEM_AI_OSCE = "ai_osce_checkbox";
     public static final String
             PREF_JOIN_ROOT_CHAN = "auto_join_root_checkbox",
             PREF_SUB_TEXTMESSAGE = "sub_txtmsg_checkbox",

Client/TeamTalkAndroid/src/main/res/values/strings.xml

@@ -163,6 +163,10 @@
     <string name="pref_summary_bluetooth_headset_checkbox">When bluetooth headset is connected use its microphone</string>
     <string name="pref_title_voice_processing_checkbox">Enable voice preprocessing</string>
     <string name="pref_summary_voice_processing_checkbox">Use echo cancellation and automatic gain control</string>
+    <string name="pref_title_ai_dred_checkbox">Enable AI packet loss recovery (DRED)</string>
+    <string name="pref_summary_ai_dred_checkbox">Make outgoing voice more resilient on unstable networks</string>
+    <string name="pref_title_ai_osce_checkbox">Enable AI speech enhancement (OSCE)</string>
+    <string name="pref_summary_ai_osce_checkbox">Improve clarity of received voice</string>
 
     <string name="text_tts_loggedin">has logged in</string>
     <string name="text_tts_loggedout">has logged out</string>

Client/TeamTalkAndroid/src/main/res/xml/pref_soundsystem.xml

@@ -19,6 +19,18 @@
         android:title="@string/pref_title_voice_processing_checkbox"
         android:summary="@string/pref_summary_voice_processing_checkbox" />
 
+    <CheckBoxPreference
+        android:key="ai_dred_checkbox"
+        android:defaultValue="false"
+        android:title="@string/pref_title_ai_dred_checkbox"
+        android:summary="@string/pref_summary_ai_dred_checkbox" />
+
+    <CheckBoxPreference
+        android:key="ai_osce_checkbox"
+        android:defaultValue="false"
+        android:title="@string/pref_title_ai_osce_checkbox"
+        android:summary="@string/pref_summary_ai_osce_checkbox" />
+
 	<CheckBoxPreference
 		android:key="bluetooth_headset_checkbox"
 		android:defaultValue="false"

Client/TeamTalkClassic/Helper.cpp

@@ -607,6 +607,11 @@ BOOL InitSoundSystem(teamtalk::ClientXML& xmlSettings, SoundDevice& indev, Sound
         effects.bEnableEchoCancellation = bEchoCancel;
     TT_SetSoundDeviceEffects(ttInst, &effects);
 
+    AIAudioEffect aiEffect = {};
+    aiEffect.bEnableDRED = xmlSettings.GetAIAudioDRED(DEFAULT_AI_DRED_ENABLE);
+    aiEffect.bEnableOSCE = xmlSettings.GetAIAudioOSCE(DEFAULT_AI_OSCE_ENABLE);
+    TT_SetAIAudioEffect(ttInst, &aiEffect);
+
     BOOL bSuccess = FALSE;
 
     if (bDuplex || (effects.bEnableEchoCancellation && (indev.nSoundSystem == SOUNDSYSTEM_WASAPI)))
@@ -644,8 +649,8 @@ extern BOOL g_bSpeech;
 
 void AddTextToSpeechMessage(const CString& szMsg)
 {
-#if defined(ENABLE_TOLK)
-    if(g_bSpeech)
-        Tolk_Output(szMsg);
+#if defined(ENABLE_TOLK)
+    if(g_bSpeech)
+        Tolk_Output(szMsg);
 #endif
 }

Client/TeamTalkClassic/TeamTalk.rc

@@ -247,12 +247,12 @@ BEGIN
     CONTROL         "",IDC_SLIDER_CLIENTSOUNDS_VOL,"msctls_trackbar32",TBS_BOTH | TBS_NOTICKS | WS_TABSTOP | 0x400,148,211,99,15
 END
 
-IDD_PROPPAGE_SOUNDSYSPAGE DIALOGEX 0, 0, 260, 221
+IDD_PROPPAGE_SOUNDSYSPAGE DIALOGEX 0, 0, 260, 249
 STYLE DS_SETFONT | DS_MODALFRAME | WS_POPUP | WS_CAPTION | WS_SYSMENU
 CAPTION "Sound System"
 FONT 8, "MS Sans Serif", 0, 0, 0x0
 BEGIN
-    GROUPBOX        "Sound system settings",IDC_STATIC_GRPSOUNDSYSTEM,7,7,246,207
+    GROUPBOX        "Sound system settings",IDC_STATIC_GRPSOUNDSYSTEM,7,7,246,235
     CONTROL         "Windows Audio Session",IDC_RADIO_WASAPI,"Button",BS_AUTORADIOBUTTON | WS_TABSTOP,41,22,95,10
     CONTROL         "DirectSound",IDC_RADIO_DIRECTSOUND,"Button",BS_AUTORADIOBUTTON | WS_TABSTOP,160,22,56,10
     LTEXT           "Input device",IDC_STATIC_INPUT,14,39,75,8
@@ -270,12 +270,16 @@ BEGIN
                     "Button",BS_AUTOCHECKBOX | WS_TABSTOP,12,136,236,10
     CONTROL         "Enable noise reduction (suppress noise from microphone)",IDC_CHECK_DENOISE,
                     "Button",BS_AUTOCHECKBOX | WS_TABSTOP,12,150,236,10
-    CONTROL         "Auto position users using 3D sound",IDC_CHECK_POSITIONING,
+    CONTROL         "Enable AI packet loss recovery (DRED)",IDC_CHECK_AI_DRED,
                     "Button",BS_AUTOCHECKBOX | WS_TABSTOP,12,164,236,10
-    LTEXT           "Media file vs. voice volume",IDC_STATIC_MEDIASTREAMVOL,12,180,104,8
-    CONTROL         "",IDC_SLIDER_MEDIASTREAM_VOL,"msctls_trackbar32",TBS_BOTH | TBS_NOTICKS | WS_TABSTOP | 0x400,117,176,131,15
-    PUSHBUTTON      "Refresh devices",IDC_BUTTON_REFRESHSND,132,193,60,12
-    PUSHBUTTON      "&Default",IDC_BUTTON_DEFAULT,197,193,51,12
+    CONTROL         "Enable AI speech enhancement (OSCE)",IDC_CHECK_AI_OSCE,
+                    "Button",BS_AUTOCHECKBOX | WS_TABSTOP,12,178,236,10
+    CONTROL         "Auto position users using 3D sound",IDC_CHECK_POSITIONING,
+                    "Button",BS_AUTOCHECKBOX | WS_TABSTOP,12,192,236,10
+    LTEXT           "Media file vs. voice volume",IDC_STATIC_MEDIASTREAMVOL,12,208,104,8
+    CONTROL         "",IDC_SLIDER_MEDIASTREAM_VOL,"msctls_trackbar32",TBS_BOTH | TBS_NOTICKS | WS_TABSTOP | 0x400,117,204,131,15
+    PUSHBUTTON      "Refresh devices",IDC_BUTTON_REFRESHSND,132,221,60,12
+    PUSHBUTTON      "&Default",IDC_BUTTON_DEFAULT,197,221,51,12
 END
 
 IDD_PROPPAGE_WINDOWPAGE DIALOGEX 0, 0, 260, 240
@@ -1083,7 +1087,7 @@ BEGIN
         RIGHTMARGIN, 253
         VERTGUIDE, 248
         TOPMARGIN, 7
-        BOTTOMMARGIN, 214
+        BOTTOMMARGIN, 242
     END
 
     IDD_PROPPAGE_WINDOWPAGE, DIALOG
@@ -1540,8 +1544,8 @@ IDR_MAINFRAME           ICON                    "res\\TeamTalk.ico"
 //
 
 VS_VERSION_INFO VERSIONINFO
- FILEVERSION 5,22,1,5199
- PRODUCTVERSION 5,22,1,5199
+ FILEVERSION 5,22,1,5199
+ PRODUCTVERSION 5,22,1,5199
  FILEFLAGSMASK 0x1fL
 #ifdef _DEBUG
  FILEFLAGS 0x1L
@@ -1558,12 +1562,12 @@ BEGIN
         BEGIN
             VALUE "CompanyName", "BearWare.dk"
             VALUE "FileDescription", "TeamTalk voice conferencing client"
-            VALUE "FileVersion", "5, 22, 1, 5199"
+            VALUE "FileVersion", "5, 22, 1, 5199"
             VALUE "InternalName", "TeamTalk5.exe"
             VALUE "LegalCopyright", "Copyright (c) 2005-2020, BearWare.dk.  All rights reserved."
             VALUE "OriginalFilename", "TeamTalk5.exe"
             VALUE "ProductName", "BearWare.dk TeamTalk"
-            VALUE "ProductVersion", "5, 22, 1, 5199"
+            VALUE "ProductVersion", "5, 22, 1, 5199"
         END
     END
     BLOCK "VarFileInfo"

Client/TeamTalkClassic/TeamTalkBase.h

@@ -95,6 +95,8 @@ AudioPreprocessor InitDefaultAudioPreprocessor(AudioPreprocessorType preprocesso
 #define DEFAULT_ECHO_ENABLE             FALSE
 #define DEFAULT_AGC_ENABLE              FALSE
 #define DEFAULT_DENOISE_ENABLE          FALSE
+#define DEFAULT_AI_DRED_ENABLE          FALSE
+#define DEFAULT_AI_OSCE_ENABLE          FALSE
 
 #define DEFAULT_SPEEXDSP_AGC_ENABLE             DEFAULT_AGC_ENABLE
 #define DEFAULT_SPEEXDSP_AGC_GAINLEVEL          8000

Client/TeamTalkClassic/TeamTalkDlg.cpp

@@ -3487,6 +3487,8 @@ void CTeamTalkDlg::OnFilePreferences()
     soundpage.m_bEchoCancel = m_xmlSettings.GetEchoCancel(DEFAULT_ECHO_ENABLE);
     soundpage.m_bAGC = m_xmlSettings.GetAGC(DEFAULT_AGC_ENABLE);
     soundpage.m_bDenoise = m_xmlSettings.GetDenoise(DEFAULT_DENOISE_ENABLE);
+    soundpage.m_bAIDRED = m_xmlSettings.GetAIAudioDRED(DEFAULT_AI_DRED_ENABLE);
+    soundpage.m_bAIOSCE = m_xmlSettings.GetAIAudioOSCE(DEFAULT_AI_OSCE_ENABLE);
     soundpage.m_nMediaVsVoice = m_xmlSettings.GetMediaStreamVsVoice(DEFAULT_MEDIA_VS_VOICE);
 
     ///////////////////////
@@ -3741,6 +3743,8 @@ void CTeamTalkDlg::OnFilePreferences()
         bRestartSound |= m_xmlSettings.GetAGC(DEFAULT_AGC_ENABLE) != bool(soundpage.m_bAGC);
         bRestartSound |= m_xmlSettings.GetEchoCancel(DEFAULT_ECHO_ENABLE) != bool(soundpage.m_bEchoCancel);
         bRestartSound |= m_xmlSettings.GetDenoise(DEFAULT_DENOISE_ENABLE) != bool(soundpage.m_bDenoise);
+        bRestartSound |= m_xmlSettings.GetAIAudioDRED(DEFAULT_AI_DRED_ENABLE) != bool(soundpage.m_bAIDRED);
+        bRestartSound |= m_xmlSettings.GetAIAudioOSCE(DEFAULT_AI_OSCE_ENABLE) != bool(soundpage.m_bAIOSCE);
 
         m_xmlSettings.SetSoundOutputDevice(soundpage.m_nOutputDevice);
         m_xmlSettings.SetSoundOutputDevice(STR_UTF8(soundpage.m_szOutputDeviceID));
@@ -3750,6 +3754,8 @@ void CTeamTalkDlg::OnFilePreferences()
         m_xmlSettings.SetEchoCancel(soundpage.m_bEchoCancel);
         m_xmlSettings.SetAGC(soundpage.m_bAGC);
         m_xmlSettings.SetDenoise(soundpage.m_bDenoise);
+        m_xmlSettings.SetAIAudioDRED(soundpage.m_bAIDRED);
+        m_xmlSettings.SetAIAudioOSCE(soundpage.m_bAIOSCE);
         if(m_xmlSettings.GetMediaStreamVsVoice(DEFAULT_MEDIA_VS_VOICE) != soundpage.m_nMediaVsVoice)
         {
             m_xmlSettings.SetMediaStreamVsVoice(soundpage.m_nMediaVsVoice);

Client/TeamTalkClassic/gui/SoundSysPage.cpp

@@ -47,6 +47,8 @@ CSoundSysPage::CSoundSysPage()
 , m_bTesting(FALSE)
 , m_bDenoise(FALSE)
 , m_bEchoCancel(DEFAULT_ECHO_ENABLE)
+, m_bAIDRED(DEFAULT_AI_DRED_ENABLE)
+, m_bAIOSCE(DEFAULT_AI_OSCE_ENABLE)
 , m_SndLoopBack(NULL)
 , m_bAGC(DEFAULT_AGC_ENABLE)
 , m_nMediaVsVoice(DEFAULT_MEDIA_VS_VOICE)
@@ -75,6 +77,10 @@ void CSoundSysPage::DoDataExchange(CDataExchange* pDX)
     DDX_Control(pDX, IDC_BUTTON_DEFAULT, m_wndDefaultBtn);
     DDX_Control(pDX, IDC_CHECK_ECHOCANCEL, m_btnEchoCancel);
     DDX_Check(pDX, IDC_CHECK_ECHOCANCEL, m_bEchoCancel);
+    DDX_Control(pDX, IDC_CHECK_AI_DRED, m_btnAIDRED);
+    DDX_Check(pDX, IDC_CHECK_AI_DRED, m_bAIDRED);
+    DDX_Control(pDX, IDC_CHECK_AI_OSCE, m_btnAIOSCE);
+    DDX_Check(pDX, IDC_CHECK_AI_OSCE, m_bAIOSCE);
     DDX_Control(pDX, IDC_STATIC_INPUT_SAMPLERATES, m_wndInputSampleRates);
     DDX_Control(pDX, IDC_STATIC_OUTPUT_SAMPLERATES, m_wndOutputSampleRates);
     DDX_Check(pDX, IDC_CHECK_AGC, m_bAGC);