Jimaku Editor for File


Stops playback when a character is entered. Deleting a character and using shortcut keys do not stop playback.
Connection Engine Name :

Speech Recognition API AmiVoice Cloud Platform is used to retrieve the results of speech recognition.
Speech recognition is performed on loaded video and audio files and converted to text. In the case of video, it converts it to an audio file in the browser and uploads it to the server.
You can use the speech recognition API for a fee. (You can try it out for 60 minutes for free every month.)
1. Go to the AmiVoice Cloud Platform page and click on "無料で利用開始" to log in. If you are already using the platform, open My Page.
2. Click "接続情報" in the menu on the left and copy the string shown in "APPKEY". This APPKEY is very important information, so please be careful with it.
3. Go back to "Jimaku Editor", go to "Speech Recognition Settings" and paste into "APPKEY", if you want to save APPKEY to your browser, please check "Save APPKEY to Browser".
4. "Connection Engine Name" setting can be changed according to your use of AmiVoice Cloud Platform. The default setting is "Conversation_General".
5. Select saved or not on the speech recognition server. Select the appropriate option according to the confidentiality of the content.
Click the "Get Speech Recognition Results" button. It may take some time depending on the length of the file, but please wait a while without closing the page. When the speech recognition is completed, the recognition result will be displayed on the right side.
Service ID
Enter the service ID to enable word registration on the AmiVoice API.
Register Words (Japanese Only)
If the destination has the same file name, the browser will automatically rename the file and save it.
Connection Engine Name
The speech recognition engine can be purchased from "プラン申込" in the "My Page" of AmiVoice Cloud Platform. Please check the price.
Option Settings
The price differs between save and not save. For details, please refer to the AmiVoice Cloud Platform My Page.
Up to 10 speakers can be identified and displayed.

Copyright © UD Talk. All Rights Reserved.