diff --git a/app/src/main/java/com/nextcloud/talk/activities/CallActivity.kt b/app/src/main/java/com/nextcloud/talk/activities/CallActivity.kt
index 7d8088e7c..003dc869a 100644
--- a/app/src/main/java/com/nextcloud/talk/activities/CallActivity.kt
+++ b/app/src/main/java/com/nextcloud/talk/activities/CallActivity.kt
@@ -940,17 +940,22 @@ class CallActivity : CallBaseActivity() {
     }
 
     private fun initSelfVideoViewForPipMode() {
-        try {
-            binding!!.pipSelfVideoRenderer.init(rootEglBase!!.eglBaseContext, null)
-        } catch (e: IllegalStateException) {
-            Log.d(TAG, "pipGroupVideoRenderer already initialized", e)
-        }
-        binding!!.pipSelfVideoRenderer.setZOrderMediaOverlay(true)
-        // disabled because it causes some devices to crash
-        binding!!.pipSelfVideoRenderer.setEnableHardwareScaler(false)
-        binding!!.pipSelfVideoRenderer.setScalingType(RendererCommon.ScalingType.SCALE_ASPECT_FIT)
+        if (!isVoiceOnlyCall) {
+            binding!!.pipSelfVideoRenderer.visibility = View.VISIBLE
+            try {
+                binding!!.pipSelfVideoRenderer.init(rootEglBase!!.eglBaseContext, null)
+            } catch (e: IllegalStateException) {
+                Log.d(TAG, "pipGroupVideoRenderer already initialized", e)
+            }
+            binding!!.pipSelfVideoRenderer.setZOrderMediaOverlay(true)
+            // disabled because it causes some devices to crash
+            binding!!.pipSelfVideoRenderer.setEnableHardwareScaler(false)
+            binding!!.pipSelfVideoRenderer.setScalingType(RendererCommon.ScalingType.SCALE_ASPECT_FIT)
 
-        localVideoTrack!!.addSink(binding!!.pipSelfVideoRenderer)
+            localVideoTrack?.addSink(binding?.pipSelfVideoRenderer)
+        } else {
+            binding!!.pipSelfVideoRenderer.visibility = View.GONE
+        }
     }
 
     private fun initGrid() {
@@ -2187,6 +2192,7 @@ class CallActivity : CallBaseActivity() {
                             startVideoCapture(true)
                         }
                     }
+
                     in ANGLE_LANDSCAPE_RIGHT_THRESHOLD_MIN..ANGLE_LANDSCAPE_RIGHT_THRESHOLD_MAX,
                     in ANGLE_LANDSCAPE_LEFT_THRESHOLD_MIN..ANGLE_LANDSCAPE_LEFT_THRESHOLD_MAX -> {
                         if (lastAspectRatio != RATIO_16_TO_9) {
@@ -2718,6 +2724,7 @@ class CallActivity : CallBaseActivity() {
                     handler!!.postDelayed({ setCallState(CallStatus.CALLING_TIMEOUT) }, CALLING_TIMEOUT)
                     handler!!.post { handleCallStateJoined() }
                 }
+
                 CallStatus.IN_CONVERSATION -> handler!!.post { handleCallStateInConversation() }
                 CallStatus.OFFLINE -> handler!!.post { handleCallStateOffline() }
                 CallStatus.LEAVING -> handler!!.post { handleCallStateLeaving() }
@@ -3241,21 +3248,30 @@ class CallActivity : CallBaseActivity() {
         binding!!.callInfosLinearLayout.visibility = View.GONE
         binding!!.selfVideoViewWrapper.visibility = View.GONE
         binding!!.callStates.callStateRelativeLayout.visibility = View.GONE
+        binding!!.pipCallConversationNameTextView.text = conversationName
 
-        binding!!.selfVideoRenderer.clearImage()
-        binding!!.selfVideoRenderer.release()
-        if (participantItems.size > 1) {
-            binding!!.pipCallConversationNameTextView.text = conversationName
-            binding!!.pipSelfVideoOverlay.visibility = View.VISIBLE
-            initSelfVideoViewForPipMode()
+        if (isVoiceOnlyCall) {
+            if (participantItems.size > 1) {
+                binding!!.pipOverlay.visibility = View.VISIBLE
+                binding!!.pipSelfVideoRenderer.visibility = View.GONE
+            } else {
+                binding!!.pipOverlay.visibility = View.GONE
+            }
         } else {
-            binding!!.pipSelfVideoOverlay.visibility = View.GONE
+            binding!!.selfVideoRenderer.clearImage()
+            binding!!.selfVideoRenderer.release()
+            if (participantItems.size > 1) {
+                binding!!.pipOverlay.visibility = View.VISIBLE
+                initSelfVideoViewForPipMode()
+            } else {
+                binding!!.pipOverlay.visibility = View.GONE
+            }
         }
     }
 
     override fun updateUiForNormalMode() {
         Log.d(TAG, "updateUiForNormalMode")
-        binding!!.pipSelfVideoOverlay.visibility = View.GONE
+        binding!!.pipOverlay.visibility = View.GONE
 
         if (isVoiceOnlyCall) {
             binding!!.callControls.visibility = View.VISIBLE
diff --git a/app/src/main/java/com/nextcloud/talk/call/components/AvatarWithFallback.kt b/app/src/main/java/com/nextcloud/talk/call/components/AvatarWithFallback.kt
index 0763df116..49a09cd6c 100644
--- a/app/src/main/java/com/nextcloud/talk/call/components/AvatarWithFallback.kt
+++ b/app/src/main/java/com/nextcloud/talk/call/components/AvatarWithFallback.kt
@@ -55,7 +55,7 @@ fun AvatarWithFallback(participant: ParticipantUiState, modifier: Modifier = Mod
                 contentAlignment = Alignment.Center
             ) {
                 Text(
-                    text = if (initials.isNotEmpty()) initials else "?",
+                    text = initials.ifEmpty { "?" },
                     color = Color.Black,
                     fontSize = 24.sp
                 )
diff --git a/app/src/main/java/com/nextcloud/talk/call/components/ParticipantGrid.kt b/app/src/main/java/com/nextcloud/talk/call/components/ParticipantGrid.kt
index 135f3e523..1833c448a 100644
--- a/app/src/main/java/com/nextcloud/talk/call/components/ParticipantGrid.kt
+++ b/app/src/main/java/com/nextcloud/talk/call/components/ParticipantGrid.kt
@@ -94,7 +94,8 @@ fun ParticipantGrid(
                 modifier = Modifier
                     .height(itemHeight)
                     .fillMaxWidth(),
-                eglBase = eglBase
+                eglBase = eglBase,
+                isVoiceOnlyCall = isVoiceOnlyCall
             )
         }
     }
@@ -261,7 +262,7 @@ fun getTestParticipants(numberOfParticipants: Int): List<ParticipantUiState> {
             sessionKey = i.toString(),
             nick = "testuser$i Test",
             isConnected = true,
-            isAudioEnabled = if (i == 3) true else false,
+            isAudioEnabled = true,
             isStreamEnabled = true,
             raisedHand = true,
             avatarUrl = "",
diff --git a/app/src/main/java/com/nextcloud/talk/call/components/ParticipantTile.kt b/app/src/main/java/com/nextcloud/talk/call/components/ParticipantTile.kt
index 35b851ea0..385927f7e 100644
--- a/app/src/main/java/com/nextcloud/talk/call/components/ParticipantTile.kt
+++ b/app/src/main/java/com/nextcloud/talk/call/components/ParticipantTile.kt
@@ -33,13 +33,18 @@ import com.nextcloud.talk.call.ParticipantUiState
 import org.webrtc.EglBase
 
 @Composable
-fun ParticipantTile(participant: ParticipantUiState, eglBase: EglBase?, modifier: Modifier = Modifier) {
+fun ParticipantTile(
+    participant: ParticipantUiState,
+    eglBase: EglBase?,
+    modifier: Modifier = Modifier,
+    isVoiceOnlyCall: Boolean,
+) {
     Box(
         modifier = modifier
             .clip(RoundedCornerShape(12.dp))
             .background(Color.DarkGray)
     ) {
-        if (participant.isStreamEnabled && participant.mediaStream != null) {
+        if (!isVoiceOnlyCall && participant.isStreamEnabled && participant.mediaStream != null) {
             WebRTCVideoView(participant, eglBase)
         } else {
             AvatarWithFallback(
@@ -113,6 +118,7 @@ fun ParticipantTilePreview() {
         modifier = Modifier
             .fillMaxWidth()
             .height(300.dp),
-        eglBase = null
+        eglBase = null,
+        isVoiceOnlyCall = false
     )
 }
diff --git a/app/src/main/res/layout/call_activity.xml b/app/src/main/res/layout/call_activity.xml
index ea97114e9..ca67beaa2 100644
--- a/app/src/main/res/layout/call_activity.xml
+++ b/app/src/main/res/layout/call_activity.xml
@@ -322,7 +322,7 @@
     </LinearLayout>
 
     <RelativeLayout
-        android:id="@+id/pipSelfVideoOverlay"
+        android:id="@+id/pipOverlay"
         android:layout_width="match_parent"
         android:layout_height="match_parent"
         android:background="@color/black"
@@ -336,9 +336,8 @@
             android:layout_width="match_parent"
             android:layout_centerHorizontal="true"
             android:layout_marginStart="5dp"
-            android:layout_marginTop="40dp"
+            android:layout_marginTop="0dp"
             android:layout_marginEnd="5dp"
-            android:layout_marginBottom="-30dp"
             android:ellipsize="end"
             android:maxLines="3"
             android:textAlignment="center"