Rimo Voice is designed so that the text color becomes lighter in areas where voice recognition has become ambiguous. Also, by linking voice data with text data, users can easily re-listen to the relevant part by clicking on the part where the text color is lighter. [Available at 20 yen for 30 seconds. What āAI transcription serviceā specialized in Japanese language aims at | DIAMOND SIGNAL https://signal.diamond.jp/articles/-/252?fbclid=IwAR2ZEiIW_B3l8pfZYLypYdk88rsHqgr7YUL9pdMp -WTLy82wuPp9d9loB94] Help
Tried.
Iām not sure if the accuracy is good or not, since I donāt usually do transcription work.
The above seems to work well because it is a normal conversation, but machine-gun talk in technical terms is naturally messy.
However, even in this state, the speaker knows which part of the story he or she is talking about, and when you click on the speaker, the playback starts from there, which makes it very easy to check the flow of the story. The playback speed can be doubled from 0.5x to 3x, although I did not use it this time because the speaker was speaking too fast.
Other Comments https://twitter.com/souta6954/status/1303127472477593600?s=21
This page is auto-translated from [/nishio/Rimo Voice](https://scrapbox.io/nishio/Rimo Voice) using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. Iām very happy to spread my thought to non-Japanese readers.