Question / Help How to convert text to speech and mix it with video?
- Thread starter obs2019
- Start date
Narcogen
Active Member
Short answer: No.
Long answer: You could recreate this but it would require manual timing of elements; breaking the text for reading and display up into individual segments, and switching from scene to scene to update these elements so that the read speech is synchronous with the displayed text. There's no way to automatically set up a text file to be read by text to speech and then also displayed on screen in segments that match the reproduction of the speech.
Long answer: You could recreate this but it would require manual timing of elements; breaking the text for reading and display up into individual segments, and switching from scene to scene to update these elements so that the read speech is synchronous with the displayed text. There's no way to automatically set up a text file to be read by text to speech and then also displayed on screen in segments that match the reproduction of the speech.
Thank you for your reply. If there is no such functionalities in OBS itself, do you have some suggestions on how to do it with OBS plugin or other tools? My plan is to find a tool that converts text to audio, then add the audio as a source in OBS. I know I can add text on video in OBS, but the text keeps the same during the video. I need another tool to convert the text into subtitle, then mix it with the video. The synchronization is not very important if difficult to implement.
Narcogen
Active Member
I'm not aware of any.
If I had to do such myself I would do it as I suggested-- breaking the text into blocks, timing them, and then using the advanced scene switcher to advance scenes based on those timings, and put individual text files and audio sources into separate scenes. If precise timing is not necessary, I would just use longer scenes.
If I had to do such myself I would do it as I suggested-- breaking the text into blocks, timing them, and then using the advanced scene switcher to advance scenes based on those timings, and put individual text files and audio sources into separate scenes. If precise timing is not necessary, I would just use longer scenes.