diff --git a/bigbluebutton-html5/imports/api/audio-captions/server/methods/updateTranscript.js b/bigbluebutton-html5/imports/api/audio-captions/server/methods/updateTranscript.js
index 3b1b7bf68c..a00802a17e 100644
--- a/bigbluebutton-html5/imports/api/audio-captions/server/methods/updateTranscript.js
+++ b/bigbluebutton-html5/imports/api/audio-captions/server/methods/updateTranscript.js
@@ -3,7 +3,7 @@ import RedisPubSub from '/imports/startup/server/redis';
 import { extractCredentials } from '/imports/api/common/server/helpers';
 import Logger from '/imports/startup/server/logger';
 
-export default function updateTranscript(transcriptId, start, end, text, transcript, locale) {
+export default function updateTranscript(transcriptId, start, end, text, transcript, locale, isFinal) {
   try {
     const REDIS_CONFIG = Meteor.settings.private.redis;
     const CHANNEL = REDIS_CONFIG.channels.toAkkaApps;
@@ -19,6 +19,7 @@ export default function updateTranscript(transcriptId, start, end, text, transcr
     check(text, String);
     check(transcript, String);
     check(locale, String);
+    check(isFinal, Boolean);
 
     // Ignore irrelevant updates
     if (start !== -1 && end !== -1) {
@@ -29,6 +30,7 @@ export default function updateTranscript(transcriptId, start, end, text, transcr
         text,
         transcript,
         locale,
+        result: isFinal,
       };
 
       RedisPubSub.publishUserMessage(CHANNEL, EVENT_NAME, meetingId, requesterUserId, payload);
diff --git a/bigbluebutton-html5/imports/ui/components/audio/captions/speech/service.js b/bigbluebutton-html5/imports/ui/components/audio/captions/speech/service.js
index c437e61c3d..dbfc908122 100644
--- a/bigbluebutton-html5/imports/ui/components/audio/captions/speech/service.js
+++ b/bigbluebutton-html5/imports/ui/components/audio/captions/speech/service.js
@@ -81,7 +81,7 @@ const initSpeechRecognition = () => {
 
 let prevId = '';
 let prevTranscript = '';
-const updateTranscript = (id, transcript, locale) => {
+const updateTranscript = (id, transcript, locale, isFinal) => {
   // If it's a new sentence
   if (id !== prevId) {
     prevId = id;
@@ -102,7 +102,7 @@ const updateTranscript = (id, transcript, locale) => {
   // Stores current transcript as previous
   prevTranscript = transcript;
 
-  makeCall('updateTranscript', id, start, end, text, transcript, locale);
+  makeCall('updateTranscript', id, start, end, text, transcript, locale, isFinal);
 };
 
 const throttledTranscriptUpdate = throttle(updateTranscript, THROTTLE_TIMEOUT, {
@@ -111,12 +111,12 @@ const throttledTranscriptUpdate = throttle(updateTranscript, THROTTLE_TIMEOUT, {
 });
 
 const updateInterimTranscript = (id, transcript, locale) => {
-  throttledTranscriptUpdate(id, transcript, locale);
+  throttledTranscriptUpdate(id, transcript, locale, false);
 };
 
 const updateFinalTranscript = (id, transcript, locale) => {
   throttledTranscriptUpdate.cancel();
-  updateTranscript(id, transcript, locale);
+  updateTranscript(id, transcript, locale, true);
 };
 
 const getSpeechLocale = (userId = Auth.userID) => {
@@ -133,7 +133,7 @@ const isLocaleValid = (locale) => LANGUAGES.includes(locale);
 
 const isEnabled = () => isLiveTranscriptionEnabled();
 
-const isWebSpeechApi = () => PROVIDER === 'webspeech' && hasSpeechRecognitionSupport() && hasSpeechLocale();
+const isWebSpeechApi = () => PROVIDER === 'webspeech';
 
 const isVosk = () => PROVIDER === 'vosk';
 
@@ -141,7 +141,7 @@ const isWhispering = () => PROVIDER === 'whisper';
 
 const isDeepSpeech = () => PROVIDER === 'deepSpeech'
 
-const isActive = () => isEnabled() && (isWebSpeechApi() || isVosk() || isWhispering() || isDeepSpeech());
+const isActive = () => isEnabled() && ((isWebSpeechApi() && hasSpeechLocale()) || isVosk() || isWhispering() || isDeepSpeech());
 
 const getStatus = () => {
   const active = isActive();