What is Speaker Progress AI Transcript Feedback?
With Speaker Progress AI Transcript Feedback, educators can enhance their Speaker Progress assignments by adding AI feedback that will analyze the qualitative aspects of a student’s speech. When creating an assignment, teachers can select the type of speech and the criteria for assessment. When it's time to grade, AI analyzes the transcript of the student’s speech and provides detailed feedback. Educators must review this feedback before it is shared with students. Educators are free to edit or delete any of the feedback items. The feedback is presented in a clear, sentence-by-sentence format, along with a comprehensive summary, helping students improve their speaking skills effectively.
What can Speaker Progress AI Transcript Feedback do?
Speaker Progress AI Transcript Feedback supports the following generative AI operations:
- Provide summary feedback on a student’s speech – Speaker Progress AI Transcript Feedback provides summary feedback for informative, persuasive, and personal presentations. It evaluates presentations on one or more of the following criteria (selected by the educator): language use, rhetorical techniques, style, organization, message clarity, quality of information, engagement strategies, emotional and intellectual appeal, and call to action.
- Provide in line feedback on a student’s speech – Speaker Progress AI Transcript Feedback provides in line feedback for student speeches based on the same criteria described above.
What is Speaker Progress AI Transcript Feedback's intended use(s)?
Speaker Progress AI Transcript Feedback is intended for educator in the loop feedback for student speeches. Any AI feedback must be reviewed by educators before it is shared with students.
Educators should not assign grades based on AI generated feedback.
What are the limitations of Speaker Progress AI Transcript Feedback? How can users minimize the impact of Speaker Progress AI Transcript Feedback's limitations when using the system?
Speaker Progress AI Transcript Feedback is not grounded in factual knowledge, so generated content should not be assumed to be factually correct. Educators should validate and review all generated information.
What are the language and locale limitations of using Speaker Progress AI feedback?
The Speaker Progress AI Transcript functionality operates by converting spoken input into a transcript. This feature currently supports U.S. English; however, it may not always ensure complete accuracy. Factors such as background noise, microphone quality, internet connectivity, as well as variations in dialect and accent, may influence the transcript's precision. Users are advised to consider these potential discrepancies, as inaccuracies in the transcript could result in corresponding inaccuracies in the generated feedback.
The system hasn't been specifically evaluated for the wide variety of dialects and sociolects represented in English.
Generative AI models are trained using vast amounts of data, and that data is most often in English. This can sometimes result in better performance in English as compared to non-English languages. As with any deployment of generative AI models, we encourage users to be mindful of the limitations of these systems for their specific use case, and cultural and linguistic contexts.
What operational factors and settings allow for effective and responsible use of Speaker Progress AI Transcript Feedback?
Speaker Progress AI Transcript Feedback is based on educator supplied inputs including speech type and evaluation criteria as well as the transcript of a student’s speech. Educators should carefully review their AI feedback settings before publishing the assignment and should be sure to review all AI feedback thoroughly before returning assignments to the student.
How can Speaker Progress AI Transcript Feedback be disabled?
Tenant admins can disable Speaker Progress AI Transcript Feedback as well as the other Teams for EDU AI features in the Teams Admin Center.