Скачать с ютуб Students can use Notta AI to get feedback on their speaking & get suggestions for improvements в хорошем качестве

Students can use Notta AI to get feedback on their speaking & get suggestions for improvements

This is a very short demonstration of how students can use Notta AI to record themselves, get a transcript, correct transcription errors, and copy the transcription into ChatGPT (or Copilot) to ask for a corrected version and then an improved version I wanted to record a shorter simulation of the complete process of a student recording himself on https://www.notta.ai/en , getting a transcript, editing it and copying it to ChatGPT (or Copilot) and asking for it to be corrected and for suggestions for improvements. In this situation the student would already have the notes they used to retell the story and the same notes and any further notes from the feedback from AI could be used to repeat the task, recording it on Notta AI https://www.notta.ai/en (obviously) and repeating the request for correction and suggestions for improvements. I checked the level of the more coherent version suggested by ChatGPT on TextInspector https://textinspector.com/ , it was B1+ 42%. This is just the right level for a student studying a course at B1 level: Overall, I feel fortunate because I have good health, a job, and food to eat every day. My family is also in good health. While I may not have a lot of money or possessions, I still consider myself lucky. ‘Overall’ is perhaps a bit ambitious for B1 as are ‘possessions’ and ‘fortunate’, I liked ‘in good health’ and the use of ‘while’ and ‘still’ to express a contrast. The video is in real time and involved recording the screen on my iPad with the microphone off and recording the audio separately on my phone using Hi-Q (free) https://play.google.com/store/apps/de... and then mixing the two files using Microsoft’s ClipChamp. This was necessary because using the microphone on my iPad to record my voice and to speak to ChatGPT and then listen to it caused a conflict that meant some parts were not recorded. One of the reasons for making the recording on my iPad was to demonstrate how to use the other combination of Speech to Text app and Gen-AI on an iPhone. This time I used Notta AI https://www.notta.ai/en combined with ChatGPT to compare that with my previous video using my Android phone, https://soundtype.ai/en and Copilot. Footnote: I wasn’t at all happy with the corrected version produced by ChatGPT: Well*, I consider *myself lucky person. Because I have health. I have a job. I have something to eat every day. All of my family is with good health. I don't know. Maybe I don't have a lot of money or I don't have a lot of things. compared to the original transcript: Well. I consider. In general. I'm lucky person. Because. I have. Because I have health. I have a job. I have something to eat every day. All my family. With a good health. I don't know. Maybe I don't have. A lot of money or I don't have a lot of things. The words marked in bold were not all corrections and I don’t like ‘I have health’ or ‘with good health’ I have made another video using exactly the same prompts that I used with ChatGPT, but this time using Copilot and Gemini, both of which made a better job of correcting the short text.    • How ChatGPT - Copilot - Gemini compar...