Accessible Audio Recording Checklist #
- Create a transcript or share the script of your audio recording
- Edit your transcript for accuracy
- Share the transcript or link to it next to relevant media
Accessible Audio Recording | The Details #
Create a Transcript #
- All audio recordings should be shared alongside a written transcript of the piece.
- If reading from a script, the script can meet the requirement as long as the recording does not deviate from script.
- If an accurate script is unavailable, the easiest method is to upload your audio to an automatic tool that can make the first pass at generating a transcript, but all transcripts must be reviewed for accuracy by a human to meet requirements.
- The browser version of Microsoft Word can be used to transcribe audio. It is free for all UMW users.
Microsoft’s Guide to Transcription in Word
Edit Your Transcript for Accuracy #
- All transcripts must be reviewed by a human to meet accessibility requirements.
- Microsoft’s transcription feature allows for review, editing, and speaker identification.
Relabeling a Speaker in Microsoft Word Transcription
Share your Transcript #
Ensure that users can easily see that a transcript is available and quickly access it. One effective approach is to place the transcript or a link to it immediately below the audio file.
Best Practices #
The primary purpose of a transcript is to provide access to information for people who cannot get it from the audio and/or video. Keeping that goal in mind will help you decide what to include and how to structure the transcript.
- Organize the content into clear paragraphs, lists, and sections. If a speaker has been talking for awhile it is acceptable to break up their speaking into multiple paragraphs if it would help with comprehension.
- Add navigation aids. Headings can make transcripts easier to use and navigate. For instance a longer presentation transcript might use section headings. It is generally acceptable to add clarifying information as long as it is clear that it was not part of the original audio.
- Include relevant non-speech information about audio. For example, if someone whispers in the audio you can denote in the transcript (whispering).
- Include timestamps only when they add value. Excessive use of timestamps can create unnecessary clutter. If you do include them, they do not need to be as detailed as caption timestamps.
- Intent over verbatim. For most transcripts content it is acceptable to prioritize capturing the intended meaning over capturing every single word. If someone is excessively using “um” or coughing throughout the transcript you can choose to omit this.
Web Accessibility Initiative (WAI) Transcripts
Web Accessibility Initiative (WAI) Transcribing Audio to Text