Skip to content

Speakr is a personal, self-hosted web application designed for transcribing audio recordings

Python

seeking

Murtaza Nasir7a91ef0c0519
diff --git a/app.py b/app.py
index 4b7cf710..805a06ff 100644
--- a/app.py
+++ b/app.py
@@ -416,7 +416,7 @@ def transcribe_audio_asr(app_context, recording_id, filepath, original_filename,
params = {
'encode': ASR_ENCODE,
'task': ASR_TASK,
- 'output': 'txt'
+ 'output': 'json'
}
if language:
params['language'] = language
@@ -435,6 +435,8 @@ def transcribe_audio_asr(app_context, recording_id, filepath, original_filename,
with httpx.Client() as client:
response = client.post(url, params=params, files=files, timeout=None)
response.raise_for_status()
+ # The ASR service now returns JSON as a string.
+ # We will store it directly. The frontend will parse it.
recording.transcription = response.text
app.logger.info(f"ASR transcription completed for recording {recording_id}.")
@@ -677,26 +679,56 @@ def update_speakers(recording_id):
if not speaker_map:
return jsonify({'error': 'No speaker map provided'}), 400
- transcription = recording.transcription
- new_participants = []
+ transcription_text = recording.transcription
+ is_json = False
+ try:
+ transcription_data = json.loads(transcription_text)
+ is_json = isinstance(transcription_data, dict) and 'segments' in transcription_data
+ except (json.JSONDecodeError, TypeError):
+ is_json = False
+
speaker_names_used = []
- for speaker_label, new_name_info in speaker_map.items():
- new_name = new_name_info['name']
- if new_name_info.get('isMe'):
- new_name = current_user.name or 'Me'
+ if is_json:
+ # Handle JSON transcript
+ for segment in transcription_data.get('segments', []):
+ original_speaker_label = segment.get('speaker')
+ if original_speaker_label in speaker_map:
+ new_name_info = speaker_map[original_speaker_label]
+ new_name = new_name_info.get('name', '').strip()
+ if new_name_info.get('isMe'):
+ new_name = current_user.name or 'Me'
+
+ if new_name:
+ segment['speaker'] = new_name
+ if new_name not in speaker_names_used:
+ speaker_names_used.append(new_name)
- if new_name:
- # Regex to handle variations like [SPEAKER_00], [ SPEAKER_00 ], etc.
- transcription = re.sub(r'\[\s*' + re.escape(speaker_label) + r'\s*\]', f'[{new_name}]', transcription)
- if new_name not in new_participants:
- new_participants.append(new_name)
- speaker_names_used.append(new_name)
-
- recording.transcription = transcription
- if new_participants:
- recording.participants = ', '.join(new_participants)
-
+ recording.transcription = json.dumps(transcription_data)
+
+ # Update participants from the final list of speakers in the JSON
+ if transcription_data.get('segments'):
+ final_speakers = set(seg.get('speaker') for seg in transcription_data['segments'] if seg.get('speaker'))
+ recording.participants = ', '.join(sorted(list(final_speakers)))
+
+ else:
+ # Handle plain text transcript
+ new_participants = []
+ for speaker_label, new_name_info in speaker_map.items():
+ new_name = new_name_info.get('name', '').strip()
+ if new_name_info.get('isMe'):
+ new_name = current_user.name or 'Me'
+
+ if new_name:
+ transcription_text = re.sub(r'\[\s*' + re.escape(speaker_label) + r'\s*\]', f'[{new_name}]', transcription_text, flags=re.IGNORECASE)
+ if new_name not in new_participants:
+ new_participants.append(new_name)
+
+ recording.transcription = transcription_text
+ if new_participants:
+ recording.participants = ', '.join(new_participants)
+ speaker_names_used = new_participants
+
# Update speaker usage statistics
if speaker_names_used:
update_speaker_usage(speaker_names_used)
diff --git a/static/js/app.js b/static/js/app.js
index ea879877..40197819 100644
--- a/static/js/app.js
+++ b/static/js/app.js
@@ -66,6 +66,7 @@ document.addEventListener('DOMContentLoaded', () => {
const isMobileUserMenuOpen = ref(false); // Mobile user icon dropdown menu state
const windowWidth = ref(window.innerWidth); // For reactive screen size
const useAsrEndpoint = ref(false);
+ const currentUserName = ref('');
// --- Audio Recording State ---
const isRecording = ref(false);
@@ -173,14 +174,35 @@ document.addEventListener('DOMContentLoaded', () => {
if (!selectedRecording.value?.transcription) {
return [];
}
- // Match both [SPEAKER_XX] and [Name] patterns
- const speakerRegex = /\[([^\]]+)\]:/g;
- const speakers = new Set();
- let match;
- while ((match = speakerRegex.exec(selectedRecording.value.transcription)) !== null) {
- speakers.add(match[1]);
+
+ const transcription = selectedRecording.value.transcription;
+ let transcriptionData;
+
+ try {
+ transcriptionData = JSON.parse(transcription);
+ } catch (e) {
+ transcriptionData = null;
+ }
+
+ if (transcriptionData && transcriptionData.segments) {
+ // JSON format - extract speakers from segments
+ const speakers = new Set();
+ transcriptionData.segments.forEach(segment => {
+ if (segment.speaker) {
+ speakers.add(segment.speaker);
+ }
+ });
+ return Array.from(speakers);
+ } else {
+ // Plain text format - use regex to find speaker patterns
+ const speakerRegex = /\[([^\]]+)\]:/g;
+ const speakers = new Set();
+ let match;
+ while ((match = speakerRegex.exec(transcription)) !== null) {
+ speakers.add(match[1]);
+ }
+ return Array.from(speakers);
}
- return Array.from(speakers);
});
const processedTranscription = computed(() => {
@@ -189,129 +211,157 @@ document.addEventListener('DOMContentLoaded', () => {
}
const transcription = selectedRecording.value.transcription;
-
- // Check if transcription has speaker labels
- const speakerRegex = /\[([^\]]+)\]:\s*/g;
- const hasDialogue = speakerRegex.test(transcription);
-
- if (!hasDialogue) {
- return {
- hasDialogue: false,
- content: transcription,
- speakers: [],
- simpleSegments: [],
- bubbleRows: []
- };
- }
+ let transcriptionData;
- // Reset regex for processing
- speakerRegex.lastIndex = 0;
-
- // Extract speakers and assign colors
- const speakers = new Set();
- let match;
- const tempTranscription = transcription;
- while ((match = speakerRegex.exec(tempTranscription)) !== null) {
- speakers.add(match[1]);
+ try {
+ transcriptionData = JSON.parse(transcription);
+ } catch (e) {
+ transcriptionData = null;
}
-
- const speakerList = Array.from(speakers);
- const speakerColors = {};
- speakerList.forEach((speaker, index) => {
- speakerColors[speaker] = `speaker-color-${(index % 8) + 1}`;
- });
- // Process transcription into dialogue segments
- const segments = [];
- const lines = transcription.split('\n');
- let currentSpeaker = null;
- let currentText = '';
+ if (transcriptionData && transcriptionData.segments) {
+ // JSON processing logic
+ const speakerColors = {};
+ const speakerList = Array.from(new Set(transcriptionData.segments.map(seg => seg.speaker)));
+ speakerList.forEach((speaker, index) => {
+ speakerColors[speaker] = `speaker-color-${(index % 8) + 1}`;
+ });
- for (const line of lines) {
- const speakerMatch = line.match(/^\[([^\]]+)\]:\s*(.*)$/);
-
- if (speakerMatch) {
- // Save previous segment if exists
- if (currentSpeaker && currentText.trim()) {
- segments.push({
- speaker: currentSpeaker,
- text: currentText.trim(),
- color: speakerColors[currentSpeaker] || 'speaker-color-1'
+ const simpleSegments = transcriptionData.segments.map(segment => ({
+ speaker: (speakerMap.value[segment.speaker]?.name || segment.speaker),
+ text: segment.text,
+ words: segment.words.map((word, index) => ({
+ ...word,
+ // Add a click handler or data attribute for seeking
+ 'data-start-time': word.start,
+ // Add space after word unless it's the last word in the segment
+ displayWord: word.word + (index < segment.words.length - 1 ? ' ' : '')
+ })),
+ color: speakerColors[segment.speaker] || 'speaker-color-1'
+ }));
+
+ // Process simple segments to add showSpeaker flag for proper display
+ const processedSimpleSegments = [];
+ let lastSpeaker = null;
+ simpleSegments.forEach(segment => {
+ processedSimpleSegments.push({
+ ...segment,
+ showSpeaker: segment.speaker !== lastSpeaker
+ });
+ lastSpeaker = segment.speaker;
+ });
+
+ const bubbleRows = [];
+ let lastBubbleSpeaker = null;
+ simpleSegments.forEach(segment => {
+ if (segment.speaker !== lastBubbleSpeaker) {
+ bubbleRows.push({
+ speaker: segment.speaker,
+ color: segment.color,
+ isMe: segment.speaker && segment.speaker.toLowerCase().includes('me'),
+ bubbles: []
});
+ lastBubbleSpeaker = segment.speaker;
}
-
- // Start new segment
- currentSpeaker = speakerMatch[1];
- currentText = speakerMatch[2];
- } else if (currentSpeaker && line.trim()) {
- // Continue current speaker's text
- currentText += ' ' + line.trim();
- } else if (!currentSpeaker && line.trim()) {
- // Text without speaker (shouldn't happen in dialogue, but handle gracefully)
- segments.push({
- speaker: null,
- text: line.trim(),
- color: 'speaker-color-1'
+ bubbleRows[bubbleRows.length - 1].bubbles.push({
+ text: segment.text,
+ words: segment.words,
+ color: segment.color
});
+ });
+
+ return {
+ hasDialogue: true,
+ isJson: true,
+ segments: simpleSegments,
+ simpleSegments: processedSimpleSegments,
+ bubbleRows: bubbleRows,
+ speakers: speakerList.map(speaker => ({
+ name: (speakerMap.value[speaker]?.name || speaker),
+ color: speakerColors[speaker] || 'speaker-color-1'
+ }))
+ };
+
+ } else {
+ // Fallback for plain text transcription
+ const speakerRegex = /\[([^\]]+)\]:\s*/g;
+ const hasDialogue = speakerRegex.test(transcription);
+
+ if (!hasDialogue) {
+ return {
+ hasDialogue: false,
+ isJson: false,
+ content: transcription,
+ speakers: [],
+ simpleSegments: [],
+ bubbleRows: []
+ };
}
- }
- // Don't forget the last segment
- if (currentSpeaker && currentText.trim()) {
- segments.push({
- speaker: currentSpeaker,
- text: currentText.trim(),
- color: speakerColors[currentSpeaker] || 'speaker-color-1'
- });
- }
+ speakerRegex.lastIndex = 0;
+ const speakers = new Set();
+ let match;
+ while ((match = speakerRegex.exec(transcription)) !== null) {
+ speakers.add(match[1]);
+ }
- // Process segments for simple view (only show speaker icon when speaker changes)
- const simpleSegments = [];
- let lastSpeaker = null;
-
- segments.forEach(segment => {
- simpleSegments.push({
- ...segment,
- showSpeaker: segment.speaker !== lastSpeaker
+ const speakerList = Array.from(speakers);
+ const speakerColors = {};
+ speakerList.forEach((speaker, index) => {
+ speakerColors[speaker] = `speaker-color-${(index % 8) + 1}`;
});
- lastSpeaker = segment.speaker;
- });
- // Process segments for bubble view (group consecutive segments by speaker)
- const bubbleRows = [];
- let currentRow = null;
-
- segments.forEach(segment => {
- // Split long text into smaller chunks for better bubble layout
- const maxChunkLength = 200; // Increased for better bundling
- const chunks = [];
-
- if (segment.text.length <= maxChunkLength) {
- chunks.push(segment.text);
- } else {
- // Split by sentences first, then by length if needed
- const sentences = segment.text.split(/(?<=[.!?])\s+/);
- let currentChunk = '';
-
- sentences.forEach(sentence => {
- if (currentChunk.length + sentence.length <= maxChunkLength) {
- currentChunk += (currentChunk ? ' ' : '') + sentence;
- } else {
- if (currentChunk) chunks.push(currentChunk);
- currentChunk = sentence;
+ const segments = [];
+ const lines = transcription.split('\n');
+ let currentSpeaker = null;
+ let currentText = '';
+
+ for (const line of lines) {
+ const speakerMatch = line.match(/^\[([^\]]+)\]:\s*(.*)$/);
+ if (speakerMatch) {
+ if (currentSpeaker && currentText.trim()) {
+ segments.push({
+ speaker: currentSpeaker,
+ text: currentText.trim(),
+ color: speakerColors[currentSpeaker] || 'speaker-color-1'
+ });
}
+ currentSpeaker = speakerMatch[1];
+ currentText = speakerMatch[2];
+ } else if (currentSpeaker && line.trim()) {
+ currentText += ' ' + line.trim();
+ } else if (!currentSpeaker && line.trim()) {
+ segments.push({
+ speaker: null,
+ text: line.trim(),
+ color: 'speaker-color-1'
+ });
+ }
+ }
+
+ if (currentSpeaker && currentText.trim()) {
+ segments.push({
+ speaker: currentSpeaker,
+ text: currentText.trim(),
+ color: speakerColors[currentSpeaker] || 'speaker-color-1'
});
-
- if (currentChunk) chunks.push(currentChunk);
}
-
- chunks.forEach((chunk, chunkIndex) => {
- // Check if we need a new row (different speaker only)
- // Remove the bubble count limit to allow better bundling
+
+ const simpleSegments = [];
+ let lastSpeaker = null;
+ segments.forEach(segment => {
+ simpleSegments.push({
+ ...segment,
+ showSpeaker: segment.speaker !== lastSpeaker
+ });
+ lastSpeaker = segment.speaker;
+ });
+
+ const bubbleRows = [];
+ let currentRow = null;
+ segments.forEach(segment => {
if (!currentRow || currentRow.speaker !== segment.speaker) {
-
if (currentRow) bubbleRows.push(currentRow);
-
currentRow = {
speaker: segment.speaker,
color: segment.color,
@@ -319,27 +369,25 @@ document.addEventListener('DOMContentLoaded', () => {
isMe: segment.speaker && segment.speaker.toLowerCase().includes('me')
};
}
-
currentRow.bubbles.push({
- text: chunk,
+ text: segment.text,
color: segment.color
});
});
- });
-
- // Don't forget the last row
- if (currentRow) bubbleRows.push(currentRow);
-
- return {
- hasDialogue: true,
- segments: segments,
- simpleSegments: simpleSegments,
- bubbleRows: bubbleRows,
- speakers: speakerList.map(speaker => ({
- name: speaker,
- color: speakerColors[speaker] || 'speaker-color-1'
- }))
- };
+ if (currentRow) bubbleRows.push(currentRow);
+
+ return {
+ hasDialogue: true,
+ isJson: false,
+ segments: segments,
+ simpleSegments: simpleSegments,
+ bubbleRows: bubbleRows,
+ speakers: speakerList.map(speaker => ({
+ name: speaker,
+ color: speakerColors[speaker] || 'speaker-color-1'
+ }))
+ };
+ }
});
const highlightedTranscript = computed(() => {
@@ -1172,6 +1220,7 @@ document.addEventListener('DOMContentLoaded', () => {
if (appDiv) {
const asrFlag = appDiv.dataset.useAsrEndpoint;
useAsrEndpoint.value = asrFlag === 'True' || asrFlag === 'true';
+ currentUserName.value = appDiv.dataset.currentUserName || '';
}
loadRecordings();
initializeDarkMode(); // Initialize dark mode on load
@@ -1473,6 +1522,14 @@ document.addEventListener('DOMContentLoaded', () => {
}
}, { deep: true });
+ watch(speakerMap, (newMap) => {
+ for (const speakerId in newMap) {
+ if (newMap[speakerId].isMe) {
+ newMap[speakerId].name = currentUserName.value || 'Me';
+ }
+ }
+ }, { deep: true });
+
// --- Reprocessing functionality ---
const confirmReprocess = (type, recording) => {
@@ -1643,15 +1700,22 @@ document.addEventListener('DOMContentLoaded', () => {
const highlightSpeakerInTranscript = (speakerId) => {
highlightedSpeaker.value = speakerId;
- nextTick(() => {
- const modalTranscript = document.querySelector('.speaker-modal-transcript');
- if (modalTranscript) {
- const firstInstance = modalTranscript.querySelector(`.speaker-tag[data-speaker-id="${speakerId}"]`);
- if (firstInstance) {
- firstInstance.scrollIntoView({ behavior: 'smooth', block: 'center' });
+ // Scroll to the first instance in the transcript when a speaker is focused
+ if (speakerId) {
+ nextTick(() => {
+ const modalTranscript = document.querySelector('div.speaker-modal-transcript');
+ if (modalTranscript) {
+ const firstInstance = modalTranscript.querySelector(`[data-speaker-id="${speakerId}"]`);
+ if (firstInstance) {
+ firstInstance.scrollIntoView({ behavior: 'smooth', block: 'center' });
+ }
}
- }
- });
+ });
+ }
+ };
+
+ const clearSpeakerHighlight = () => {
+ highlightedSpeaker.value = null;
};
const autoIdentifySpeakers = async () => {
@@ -2093,6 +2157,44 @@ document.addEventListener('DOMContentLoaded', () => {
transcriptionViewMode.value = transcriptionViewMode.value === 'simple' ? 'bubble' : 'simple';
};
+ const seekAudio = (time, context = 'main') => {
+ let audioPlayer = null;
+ if (context === 'modal') {
+ // The audio player in the modal has the class directly on the audio element
+ audioPlayer = document.querySelector('audio.speaker-modal-transcript');
+ } else {
+ audioPlayer = document.querySelector('.main-content-area audio');
+ }
+
+ if (audioPlayer) {
+ audioPlayer.currentTime = time;
+ audioPlayer.play();
+ } else {
+ console.warn(`Audio player not found for context: ${context}`);
+ // Fallback to old method if new one fails
+ const oldPlayer = document.querySelector('audio');
+ if(oldPlayer) {
+ oldPlayer.currentTime = time;
+ oldPlayer.play();
+ }
+ }
+ };
+
+ const seekAudioFromEvent = (event) => {
+ const wordSpan = event.target.closest('[data-start-time]');
+ if (!wordSpan) return;
+
+ const time = parseFloat(wordSpan.dataset.startTime);
+ if (isNaN(time)) return;
+
+ // Determine context by checking if we're inside the speaker modal
+ // The speaker modal has a specific structure with showSpeakerModal condition
+ const isInSpeakerModal = event.target.closest('.speaker-modal-transcript') !== null;
+ const context = isInSpeakerModal ? 'modal' : 'main';
+
+ seekAudio(time, context);
+ };
+
return {
// State
currentView, dragover, recordings, selectedRecording, // currentRecording removed
@@ -2161,7 +2263,9 @@ document.addEventListener('DOMContentLoaded', () => {
closeSpeakerModal,
saveSpeakerNames,
highlightedTranscript,
+ highlightedSpeaker,
highlightSpeakerInTranscript,
+ clearSpeakerHighlight,
useAsrEndpoint,
// Mobile tabs
selectedMobileTab,
@@ -2178,7 +2282,9 @@ document.addEventListener('DOMContentLoaded', () => {
autoIdentifySpeakers,
isAutoIdentifying,
// Main column resizer refs (not needed in template but good practice if they were)
- // leftMainColumn, rightMainColumn, mainColumnResizer, mainContentColumns
+ // leftMainColumn, rightMainColumn, mainColumnResizer, mainContentColumns,
+ seekAudio,
+ seekAudioFromEvent
}
},
delimiters: ['${', '}'] // Keep Vue delimiters distinct from Flask's Jinja
diff --git a/templates/index.html b/templates/index.html
index 25f4ee12..318d91c5 100644
--- a/templates/index.html
+++ b/templates/index.html
@@ -18,7 +18,8 @@
id="app"
class="container mx-auto px-4 sm:px-6 lg:px-8 py-6 flex flex-col"
:class="{ 'use-asr-endpoint': useAsrEndpoint }"
- data-use-asr-endpoint="{{ use_asr_endpoint }}">
+ data-use-asr-endpoint="{{ use_asr_endpoint }}"
+ data-current-user-name="{{ current_user.name or '' }}">
<header class="flex justify-between items-center mb-6 pb-4 border-b border-[var(--border-primary)]">
<h1 class="text-3xl font-bold text-[var(--text-primary)] cursor-pointer flex items-center" @click="switchToGalleryView" title="Go to Gallery">
<img src="{{ url_for('static', filename='img/favicon.svg') }}" alt="Speakr Logo" class="h-9 w-9 mr-2">
@@ -263,7 +264,7 @@
</span>
</div>
<label class="flex items-center text-sm text-[var(--text-muted)]">
- <input type="checkbox" v-model="speakerMap[speaker].isMe" @focus="highlightSpeakerInTranscript(speaker)" class="h-4 w-4 rounded border-gray-300 text-blue-600 focus:ring-blue-500">
+ <input type="checkbox" v-model="speakerMap[speaker].isMe" @focus="highlightSpeakerInTranscript(speaker)" @blur="clearSpeakerHighlight" class="h-4 w-4 rounded border-gray-300 text-blue-600 focus:ring-blue-500">
<span class="ml-2">Me</span>
</label>
</div>
@@ -275,7 +276,7 @@
v-model="speakerMap[speaker].name"
@input="searchSpeakers($event.target.value, speaker)"
@focus="highlightSpeakerInTranscript(speaker)"
- @blur="setTimeout(() => speakerSuggestions[speaker] = [], 200)"
+ @blur="clearSpeakerHighlight(); setTimeout(() => speakerSuggestions[speaker] = [], 200)"
:disabled="speakerMap[speaker].isMe"
class="w-full px-3 py-2 border border-[var(--border-secondary)] rounded-md shadow-sm focus:outline-none focus:ring-[var(--border-focus)] focus:border-[var(--border-focus)] sm:text-sm bg-[var(--bg-input)] text-[var(--text-primary)] disabled:bg-[var(--bg-tertiary)] disabled:text-[var(--text-muted)] disabled:cursor-not-allowed"
:placeholder="speakerMap[speaker].isMe ? 'Me' : 'Enter name for ' + speaker"
@@ -315,14 +316,27 @@
<div class="mb-4 flex-shrink-0">
<h4 class="font-semibold text-[var(--text-secondary)] mb-2">Audio Player</h4>
<div class="bg-[var(--bg-audio-player)] p-3 rounded-lg shadow border border-[var(--border-primary)]">
- <audio controls class="w-full h-10" :key="selectedRecording.id" :src="'/audio/' + selectedRecording.id">
+ <audio controls class="w-full h-10 speaker-modal-transcript" :key="selectedRecording.id" :src="'/audio/' + selectedRecording.id">
Your browser does not support the audio element.
</audio>
</div>
</div>
<!-- Transcript Section -->
- <div class="flex-grow overflow-y-auto custom-scrollbar speaker-modal-transcript" v-html="highlightedTranscript">
+ <div class="flex-grow overflow-y-auto custom-scrollbar speaker-modal-transcript">
+ <div v-if="processedTranscription.isJson">
+ <div v-for="(segment, index) in processedTranscription.simpleSegments" :key="index" class="speaker-segment mb-2">
+ <div v-if="segment.showSpeaker"
+ :class="[segment.color, 'speaker-tablet', 'font-semibold', 'mb-1', { 'speaker-highlight': highlightedSpeaker === segment.speaker }]"
+ :data-speaker-id="segment.speaker">
+ ${ segment.speaker }
+ </div>
+ <div class="speaker-text" @click="seekAudioFromEvent">
+ <span v-for="word in segment.words" :key="word.start" class="word cursor-pointer hover:bg-[var(--bg-accent)] hover:text-[var(--text-accent)] rounded px-0.5" :data-start-time="word.start">${ word.displayWord }</span>
+ </div>
+ </div>
+ </div>
+ <div v-else v-html="highlightedTranscript" @click="seekAudioFromEvent"></div>
</div>
</div>
</div>
@@ -523,17 +537,26 @@
<!-- Transcription Content -->
<div v-if="processedTranscription.hasDialogue">
<!-- Simple View -->
- <div v-if="transcriptionViewMode === 'simple'" class="transcription-simple-view">
+ <div v-if="transcriptionViewMode === 'simple'" class="transcription-simple-view" @click="seekAudioFromEvent">
<div v-for="(segment, index) in processedTranscription.simpleSegments" :key="index" class="speaker-segment">
<div v-if="segment.showSpeaker" :class="[segment.color, 'speaker-tablet']">${ segment.speaker }</div>
- <div class="speaker-text">${ segment.text }</div>
+ <div class="speaker-text">
+ <span v-if="processedTranscription.isJson" v-for="word in segment.words" :key="word.start" class="word" :data-start-time="word.start">${ word.displayWord }</span>
+ <span v-else>${ segment.text }</span>
+ </div>
</div>
</div>
<!-- Bubble View -->
- <div v-else class="transcription-with-speakers">
- <div v-for="(segment, index) in processedTranscription.segments" :key="index"
- :class="[segment.color, 'speaker-bubble', { 'speaker-me': segment.speaker && segment.speaker.toLowerCase().includes('me') }]">
- <div class="speaker-bubble-content">${ segment.text }</div>
+ <div v-else class="transcription-with-speakers" @click="seekAudioFromEvent">
+ <div v-for="(row, index) in processedTranscription.bubbleRows" :key="index"
+ :class="['bubble-row', { 'speaker-me': row.isMe }]">
+ <div v-for="(bubble, bubbleIndex) in row.bubbles" :key="bubbleIndex"
+ :class="[row.color, 'speaker-bubble', { 'speaker-me': row.isMe }]">
+ <div class="speaker-bubble-content">
+ <span v-if="processedTranscription.isJson" v-for="word in bubble.words" :key="word.start" class="word" :data-start-time="word.start">${ word.displayWord }</span>
+ <span v-else>${ bubble.text }</span>
+ </div>
+ </div>
</div>
</div>
</div>
@@ -961,21 +984,27 @@
</div>
<!-- Transcription Content -->
- <div v-if="processedTranscription.hasDialogue">
+ <div v-if="processedTranscription.hasDialogue" @click="seekAudioFromEvent">
<!-- Simple View -->
<div v-if="transcriptionViewMode === 'simple'" class="transcription-simple-view">
<div v-for="(segment, index) in processedTranscription.simpleSegments" :key="index" class="speaker-segment">
<div v-if="segment.showSpeaker" :class="[segment.color, 'speaker-tablet']">${ segment.speaker }</div>
- <div class="speaker-text">${ segment.text }</div>
+ <div class="speaker-text">
+ <span v-if="processedTranscription.isJson" v-for="word in segment.words" :key="word.start" class="word" :data-start-time="word.start">${ word.displayWord }</span>
+ <span v-else>${ segment.text }</span>
+ </div>
</div>
</div>
<!-- Bubble View -->
<div v-else class="transcription-with-speakers">
- <div v-for="(row, index) in processedTranscription.bubbleRows" :key="index"
+ <div v-for="(row, index) in processedTranscription.bubbleRows" :key="index"
:class="['bubble-row', { 'speaker-me': row.isMe }]">
- <div v-for="(bubble, bubbleIndex) in row.bubbles" :key="bubbleIndex"
+ <div v-for="(bubble, bubbleIndex) in row.bubbles" :key="bubbleIndex"
:class="[row.color, 'speaker-bubble', { 'speaker-me': row.isMe }]">
- <div class="speaker-bubble-content">${ bubble.text }</div>
+ <div class="speaker-bubble-content">
+ <span v-if="processedTranscription.isJson" v-for="word in bubble.words" :key="word.start" class="word" :data-start-time="word.start">${ word.displayWord }</span>
+ <span v-else>${ bubble.text }</span>
+ </div>
</div>
</div>
</div>