Recently, not a day goes by without seeing terms such as ChatGPT and generative AI in newspapers and on TV. I wonder what the fuss is about. The magazine "Weekly Toyo Keizai" also has a feature article with the exaggerated name "ChatGPT Work Technique Revolution".
The article introduced a tool called CLOVA Note, which transcribes the audio of meetings. In a previous post, I wrote that Word's transcription capabilities are still in the early stages of development. I tried using the same sound source to see how CLOVA Note could convert it.
It seems that this software inserts "Participant 1 00:03" in the middle when the speech is interrupted. The part in bold is different from the original text. However, the number of them is small, and you can batch convert them as text later.
Also, in the case of Word's transcription function, immediate conversion of microphone voice is free, but conversion of recorded voice data requires a paid Microsoft product. However, CLOVA Note can transcribe audio data such as recorded M4A format for free if you have a LINE account.
If the draft is completed to this extent, I think that there will be fewer parts to be reworked by people. A person was impressed with this tool, as you can see this.