Multilingual Voiceover Assistant


The Multilingual Voiceover Assistant is a cohesive tool for creating voiceovers and subtitles for video presentations and screencasts.
Utilizing advanced text-to-speech technology, it produces natural sounding voices in sync with subtitles.
This feature lets you reuse video footage by simply overlaying it with various translated voiceovers and subtitles.
The app includes straightforward video INSTRUCTIONS to assist with its use.



Step 1: Choose original language



Step 2: Write text script

                            
It's useful to start a video script with some extra lines that will be removed later, to help with synchronization with the video's start.
For instance, you could include a countdown like:

3
2
1

Always insert pauses between the lines like in the line below:
#2000
They assist you in keeping pace with the voice-over during video recording.
Unnecessary pauses can be easily removed from the completed video in any video editor.
To insert pauses between lines, write a code of form #2000 on a separate line. The number represents the pause duration in milliseconds.
In this way, #2000 corresponds to a 2-second pause. #3500 would be 3.5-second pause.
You can specify the exact time for any line to kick in.
For that, you use codes such as "#00:05:30,500" on a separate line.
This indicates that the next line will start precisely 5 minutes and 30.5 seconds from the beginning of the video.

Use Speech Synthesis Markup Language (SSML) when necessary.
It helps in pronouncing abbreviations, speaking words in a different language, or creating a pause within a sentence.
SSML tags in the application correspond to what AWS Amazon Polly uses.
For full instructions, please visit their website:
Supported SSML Tags

See some examples:

Hello there.<break time="2s"/> How are you today?
I can speak French. <lang xml:lang="fr-FR">Je ne parle pas français.</lang>
You say, <phoneme alphabet="ipa" ph="pɪˈkɑːn">pecan</phoneme>.
I say, <phoneme alphabet="ipa" ph="ˈpi.kæn">pecan</phoneme>.
2 devided by 9 is pronounced <say-as interpret-as="fraction">2/9</say-as>.
<say-as interpret-as="characters">ICP</say-as> is the abbreviation for Infineon Cloud Platform.
                            
                        


Step 3: Add translations

Step 4: Select voices

Step 5: Make voiceover and subtitles