Skip to main content

Label Speakers

After automatic diarization, speakers are assigned identifiers that you can rename to actual names. This guide shows you how to manage speaker labels throughout your transcript using the speaker dropdown menu.

Why Label Speakers?​

Benefits of Proper Speaker Labels:

  • Clarity - Know who said what at a glance
  • Professionalism - Proper names in exported transcripts
  • Searchability - Find all utterances by a specific person
  • Context - Understand conversations better
  • Distribution - Share transcripts with meaningful speaker identification

Generic vs Named Labels:

Before:
[00:00:05] Speaker 1
Thank you for joining us today.

[00:00:15] Speaker 2
Happy to be here.

After:
[00:00:05] John Smith
Thank you for joining us today.

[00:00:15] Sarah Johnson
Happy to be here.

Speaker Dropdown Menu​

Each utterance has a speaker field with a dropdown menu for managing speaker labels.

Accessing the Speaker Menu​

Location: Beginning of each utterance row, before the timestamp

Visual Appearance:

  • Badge with blue dot icon and speaker name
  • Ring border (blue-200)
  • Rounded pill shape
  • Clickable area

How to Open:

  1. Click on the speaker badge (e.g., "Speaker 1")
  2. Dropdown menu appears anchored below the button
  3. Menu shows all speaker management options

Technical: Uses Headless UI Menu component with Portal rendering for proper z-index stacking.

Speaker Menu Structure​

The dropdown menu contains these sections (in order):

┌────────────────────────────┐
│ ✓ Speaker 1                │ ← Current speaker (checkmark)
│ ○ Speaker 2                │ ← Other speakers (circle icon)
│ ○ Speaker 3                │
├────────────────────────────┤
│ ✏ Rename this speaker      │ ← Inline rename (edit icon)
│   [input field + Save]     │   (appears when clicked)
├────────────────────────────┤
│ + Add speaker              │ ← Create new speaker
│ ✖ No speaker               │ ← Remove speaker label
│ − No name                  │ ← Set to "No name"
└────────────────────────────┘

Menu Features:

  • Max height: 60vh (scrollable for many speakers)
  • Sorted speaker list (numeric defaults first, then alphabetical)
  • Visual indicators (checkmark for current, circle for others)
  • Hover effects (gray-50 background)

Renaming Speakers​

Inline Rename (Primary Method)​

Step-by-Step:

  1. Click speaker badge on any utterance
  2. Dropdown menu opens
  3. Click "Rename this speaker" (✏ Edit icon)
  4. Inline input field appears in the menu
  5. Type the new name (e.g., "John Smith")
  6. Click "Save" button or press Enter
  7. All utterances with that speaker update automatically

UI Behavior:

  • Input field auto-focuses and auto-selects existing text
  • Placeholder: "Enter new name"
  • Save button appears next to input
  • Press Escape to cancel rename

Example:

Before: All utterances labeled "Speaker 1"
Action: Click "Speaker 1" → Rename this speaker → Type "John Smith" → Save
After: All "Speaker 1" utterances now show "John Smith"

Important Implementation Details:

  • Renaming affects ALL utterances with that speaker (replaceSpeakers() function)
  • If new name matches an existing speaker, merges into that speaker
  • Case-insensitive matching (e.g., "john smith" merges with "John Smith")
  • Empty name is ignored (rename cancelled)

Character Limits​

Maximum Speaker Name Length: 40 characters

Enforced At:

  • Add Speaker modal: maxLength={40}
  • Character counter shown: "X/40"
  • Input truncates at limit

Best Practice: Keep names concise for readability in the UI.

Naming Best Practices​

Use Full Names:

  • "John Smith" rather than "John"
  • Helps distinguish multiple people with same first name
  • More professional for official transcripts

Add Role/Title (Optional):

  • "Dr. Sarah Johnson"
  • "Mike Chen (CEO)"
  • "Interviewer"
  • Useful for context in complex discussions

Be Consistent:

  • Use the same format for all speakers
  • Either all first names or all full names
  • Either all with titles or all without
  • Consistency improves readability

Examples of Good Speaker Labels:

  • "Alex Rodriguez"
  • "Dr. Sarah Kim"
  • "Jane Doe (Interviewer)"
  • "Customer"
  • "Support Agent"
  • "Host"

Changing Speaker Attribution​

Selecting Different Speaker​

When to Use:

  • Utterance incorrectly attributed to wrong speaker
  • Diarization made an error
  • Need to correct speaker assignment

How to Change:

  1. Click the speaker badge on the utterance to change
  2. Dropdown menu opens showing all speakers
  3. Click the correct speaker from the list
  4. Menu closes
  5. Utterance immediately updates to new speaker

Visual Indicators:

  • Current speaker has checkmark icon (✓)
  • Other speakers have circle-minus icon (○)
  • Active speaker also shows small checkmark on right side

Example:

Before:
[00:01:15] Speaker 1
Actually, I disagree with that point.

Realize: This was Speaker 2, not Speaker 1

Action: Click "Speaker 1" → Click "Speaker 2" from list

After:
[00:01:15] Speaker 2
Actually, I disagree with that point.

Adding New Speakers​

Add Speaker Modal​

When to Use:

  • Need a speaker not in the current list
  • Adding someone who wasn't automatically detected
  • Creating a new speaker category

How to Add:

  1. Click speaker badge on the utterance
  2. Dropdown menu opens
  3. Click "Add speaker" (+ icon)
  4. Modal dialog opens:
    • Title: "New speaker"
    • Caption: "Add new speaker in the list"
    • Input field with placeholder "Speaker 1"
    • Character counter (X/40)
  5. Type new speaker name
  6. Click "Add speaker" button (disabled if empty)
  7. Modal closes
  8. Utterance assigned to new speaker
  9. New speaker now available in all dropdown menus

Technical Details:

  • Component: AddSpeaker modal
  • Input auto-focuses on open
  • Trim whitespace before saving
  • Empty names ignored

Example:

Current speakers: Speaker 1, Speaker 2
Action: Add speaker → Type "Michael Brown" → Add speaker
New speakers: Speaker 1, Speaker 2, Michael Brown
Utterance assigned to: Michael Brown

Special Speaker Options​

No Speaker​

Button: "No speaker" with ✖ icon (UserX)

What It Does:

  • Removes speaker label from utterance
  • Sets speaker field to empty string ('')
  • Utterance displays without speaker attribution

When to Use:

  • Utterance doesn't need speaker attribution
  • Non-speech content (music, silence, sound effects)
  • Uncertain who the speaker is

Example:

Before:
[00:02:30] Speaker 1
[Pause, background music]

After selecting "No speaker":
[00:02:30]
[Pause, background music]

No Name​

Button: "No name" with − icon (Minus)

What It Does:

  • Sets speaker label to "No name"
  • Utterance displays "No name" as speaker
  • Treated as a speaker category

When to Use:

  • Placeholder for speakers you'll identify later
  • Group unnamed speakers
  • Temporary labeling

Sorting Behavior:

  • "No name" speakers sort to the bottom of the speaker list
  • Empty strings also sort to bottom

Speaker List Management​

Speaker Sorting​

Sort Order (automatically applied):

  1. Empty/No name - Bottom of list
  2. Numeric speakers - "Speaker 1", "Speaker 2", "Speaker 3", etc. (sorted numerically)
  3. Plain numbers - "1", "2", "3" (if present)
  4. Named speakers - Alphabetical with case-insensitive, natural sort

Implementation:

const parseNumericDefault = (s: string): number | null => {
  const t = s.trim();
  const m1 = t.match(/^speaker\s*(\d+)$/i);  // "Speaker 1"
  if (m1) return Number(m1[1]);
  const m2 = t.match(/^\d+$/);  // "1"
  if (m2) return Number(t);
  return null;
};

Result: Speakers appear in logical order: system-assigned numeric labels first, then custom names alphabetically.

Speaker List Updates​

Automatic Updates:

  • Speaker list derived from all utterances in document
  • Adding speaker via modal adds to list immediately
  • Renaming speaker updates all dropdown menus
  • Removing all instances of a speaker removes from list

Speaker Deduplication:

  • Case-insensitive matching
  • "john smith" and "John Smith" treated as same speaker
  • Renaming to existing speaker merges them

Utterance Actions (Adjacent to Speaker)​

The speaker row includes action buttons on the right:

Button Order (left to right):

  1. Add Utterance Above (only on first utterance)
  2. Add Utterance Below (down arrow)
  3. Split Utterance (scissors icon)
  4. Delete Utterance (X)

Visibility:

  • Buttons only enabled when utterance is active (currently focused)
  • Subtle blue hint on active row
  • Prevents accidental clicks

See Also: Merge & Split Speakers for splitting and managing utterances.

Speaker Workflow Strategies​

Workflow 1: Rename All First​

Best For: Transcripts where diarization is accurate

Steps:

  1. Open transcript after diarization completes
  2. Identify all unique speakers (Speaker 1, 2, 3, etc.)
  3. For each speaker ID:
    • Sample a few utterances (click to hear audio)
    • Identify who they are
    • Rename via inline rename
  4. Once all renamed, proceed with text editing

Benefits:

  • Get oriented with who's who first
  • Easier to read transcript during editing
  • Single-pass speaker renaming

Workflow 2: Rename as You Edit​

Best For: Long transcripts, unfamiliar speakers

Steps:

  1. Start editing from the beginning
  2. When you encounter a speaker:
    • Listen to identify who they are
    • Rename the speaker inline
  3. Continue editing
  4. Already-renamed speakers don't need attention

Benefits:

  • Verify speaker identity with audio context
  • Don't need to identify all speakers upfront
  • Incremental progress

Workflow 3: Merge via Rename​

Best For: Transcripts with over-segmentation (same person has multiple IDs)

Steps:

  1. Identify duplicate speaker IDs (e.g., "Speaker 1" and "Speaker 4" are same person)
  2. Sample utterances to confirm
  3. Rename one to match the other (e.g., rename "Speaker 4" to "Speaker 1")
  4. System merges automatically (case-insensitive matching)
  5. Once consolidated, rename to actual name
  6. Proceed with text editing

Benefits:

  • Simplifies speaker management
  • Reduces clutter in speaker list
  • More accurate transcript structure

Troubleshooting​

Speaker Dropdown Won't Open​

Issue: Clicking speaker badge doesn't open menu

Possible Causes:

  • Browser issue
  • JavaScript error
  • Focus conflict

Solutions:

  1. Click outside the utterance first
  2. Refresh the page
  3. Check browser console for errors (F12)
  4. Try different browser

Rename Doesn't Save​

Issue: Type new name but it doesn't persist

Possible Causes:

  • Name is empty after trimming whitespace
  • Pressed Escape instead of Enter/Save
  • Connection issue

Solutions:

  1. Ensure name is not just spaces
  2. Click "Save" button or press Enter
  3. Don't press Escape (cancels rename)
  4. Check internet connection

All Utterances Changed Unexpectedly​

Issue: Renamed one speaker but many utterances changed

Expected Behavior: This is correct! Inline rename replaces ALL instances of that speaker.

Understanding:

  • Inline rename is a bulk operation
  • All utterances with that speaker ID update
  • This is intentional for efficiency

If Unintended:

  • Use Undo (Ctrl/Cmd + Z)
  • To change only one utterance: Select a different speaker from the dropdown instead of renaming

Common Scenarios​

Scenario 1: Two-Person Interview​

Setup:

  • Speaker 1: Interviewer
  • Speaker 2: Guest

Steps:

  1. Click any "Speaker 1" → Rename → Type "Interviewer" → Save
  2. Click any "Speaker 2" → Rename → Type "Jane Doe" → Save
  3. Done - all utterances now properly labeled

Scenario 2: Meeting with Many Speakers​

Setup:

  • 6 speakers detected
  • Some talk frequently, others rarely

Steps:

  1. Rename frequent speakers first (most utterances)
  2. Leave infrequent speakers as "Speaker X" initially
  3. Rename as encountered during editing
  4. Sample utterances if unsure of identity

Scenario 3: Over-Segmentation (Same Person, Multiple IDs)​

Setup:

  • "Speaker 1" and "Speaker 3" are both John
  • Diarization split John into two IDs

Steps:

  1. Sample utterances to confirm both are John
  2. Click any "Speaker 3" → Rename → Type "Speaker 1" → Save
  3. All "Speaker 3" becomes "Speaker 1" (merged)
  4. Click any "Speaker 1" → Rename → Type "John Smith" → Save
  5. All John's utterances now labeled "John Smith"

Frequently Asked Questions​

Do speaker names appear in exports?​

Yes, speaker names are included in most export formats (DOCX, TXT, SRT, etc.). Format depends on export settings.

Can I use special characters in speaker names?​

Most characters work, but the 40-character limit applies. Avoid characters that might cause export issues (e.g., newlines, control characters).

How do I change just one utterance's speaker?​

Click the speaker badge and select a different speaker from the dropdown list. Don't use "Rename this speaker" (that's bulk).

Can I rename speakers after exporting?​

Exported files are separate from Scriptix. Rename speakers in Scriptix before exporting, or edit the exported file manually.

What happens if I rename to an existing speaker name?​

The system merges them automatically (case-insensitive). All utterances from both speakers combine under one name.

Can I delete a speaker from the list?​

Speakers are automatically removed when no utterances use them. Reassign all their utterances to remove a speaker from the list.

Does renaming affect the audio?​

No, speaker labels are metadata in the transcript only. Audio files are never modified.

Next Steps​

After labeling speakers:

Name your speakers clearly! Proper labels make transcripts readable and professional.