Important Notes
The feature works perfectly when there is only one speaker.
For two or more speakers, some manual labeling is still required.
You can always undo changes by clicking the UNDO button in the top navigation bar.
Step-by-Step Instructions
In the transcript, add the speaker's name to any one exchange, and click ‘Save’
Go to the “Speakers Breakdown” option under the “File” menu in the top navigation bar.
Click “Label Remaining” to auto-fill the rest of the exchanges.
For Two-Speaker Files:
Manually label all exchanges for Speaker 1.
Label one exchange for Speaker 2.
Go to “Speakers Breakdown” under the “File” menu.
Click “Label Remaining” for Speaker 2 to complete the labeling.
For Files with More Than Two Speakers:
Manually label all exchanges except one speaker.
Label one exchange for the final speaker.
Navigate to “Speakers Breakdown” in the “File” menu.
Click “Label Remaining” for the last speaker.
Common Issues and Troubleshooting
Auto-Fill labeled speakers incorrectly. What do I do?
Click the UNDO button in the top navigation to revert changes.
Why isn’t Auto-Fill labeling as expected?
Ensure you’ve manually labeled enough exchanges to guide the tool.
Frequently Asked Questions
Can Speaker Auto-Fill perfectly identify multiple speakers?
Not yet. While the feature is improving, manual labeling is still recommended for files with two or more speakers.
How can I make the process faster for files with many speakers?
Label the speakers with the fewest exchanges first, then use “Label Remaining” for the speaker with the most exchanges to save time.
For further assistance, please contact Sonix support at [email protected] or through the chatbox located in the bottom right corner of our website.