Once a file is transcribed, you can view, edit, and download the transcript from the file’s Media Details page. Files processed to include speaker separation produce a transcript that can be viewed in two different formats — one that captures the transcribed text in a single-paragraph and a second that organizes the speech of individual speakers into separate paragraphs.
To view a transcript with speaker separation:
- From the CMS homepage, find the file and click on the filename. The Media Details page opens.
2. The transcript for the engine shown displays. If more than one engine transcribed the file, click the dropdown to select a different engine and view the corresponding transcript.
3. After selecting an engine, choose a format to view the transcript.
- Speaker Separation: Includes a paragraph break each time speaker changes.
- Transcript: Provides the full transcription text in one continuous paragraph.
4. Review the transcription results.
a. The embedded media player synchronizes speech in the video/audio file with the text in the transcript, making it easy to play back the file, listen to what was said, and compare it to the transcription text.
b. A speaker label is assigned to each voice in the conversation to help identify who is talking. A blue speaker label indicates who is currently speaking and where the audio playback falls in the transcript.
c. In addition to providing the text of what a person said, each paragraph is marked with a timestamp for the precise start/end times that the words were spoken in the conversation.
d. While in playback mode, the word being spoken is highlighted in the transcription text, making it easier to follow along and proofread. The media player and the transcript text are mutually responsive. Click any word in the transcript to start video/audio playback from the selected word.
To make adjustments to the speaker separation output or transcript text, follow the steps in How to Edit