Use a person's recording
Have a real person read the script in one take, and Showreel Studio splits it into lines for the film.
Instead of a generated voice, a person can read the script themselves. They record the whole voice-over in one take, and Showreel Studio splits it back into the film's lines at the pauses, just as it does with a generated read.
Before you start
- The storyboard's voice-over final: Edit a storyboard. The recording has to match the lines.
- A recording of the whole script as an audio file, such as a
.wav.
Record the read
- Read every line's voice-over in order, from the first line to the last. Chapter cards have no words, so skip them.
- Read it in one continuous take, in the same tone throughout.
- Leave a clear pause, about two seconds, between one line and the next. The split finds the lines by these pauses.
- Keep other pauses short, so they are not mistaken for the end of a line.
- Save it as an audio file.
Use it
In the terminal
Voice the film from the person's recording:
showreel voice compose --real take.wavThe recording is copied beside the generated takes (never over them), split into lines and trimmed. It keeps the person's own tempo. To have it paced to the kit's target like a generated read, add --pace:
showreel voice compose --real take.wav --paceOnce it is voiced, the Voice panel in the app shows it as the voice in the film, and you record and build as usual.
If it will not split
Voicing stops with a message when the recording can't be split into exactly the film's lines. Usually a pause between two lines was too short, or there was a long pause in the middle of a line. Record that stretch again with a clearer stop, or re-read the whole script.