Project Overview
In the past few weeks, I was subtitling and dubbing the video “Mirabel and Isabela Argument Scene” from Disney’s film Encanto. I chose this video as my project to work on because this is one of my favorite movies in 2022. Encanto tells a story of a family, the Madrigals, that all the children in the family have been blessed with a gift except Mirabel.
The video I chose to localize was a scene in which Mirabel argues with her elder sister Isabela. The source language of the video was English, and I made subtitles and dubbed the video into Simplified Chinese.
Tools I Used
Adobe Audition: Edit the sound files
Adobe Premiere Pro: Edit + subtitling the video
Downie 4: Download the source video from YouTube
iPhone: Record vocals in the target language
Moises: Separate vocals and background music from the source video
Process & Challenges
File Preparation
First, I downloaded the video from YouTube using Downie 4, a YouTube Video Downloader for macOS, and it was super easy to use (not free though :-0).
After checking the source video worked well, I imported the file into Moises and extracted vocals and background music from the video.
Based on my localization experience, file preparation is the most important step of a localization workflow. It is where a stakeholder can get to know whether the source assets received from the client run successfully or pass localizability checks. Otherwise, the localization team may need to recreate the source assets or ask the client for another successful version.
Translation
No CAT tools were used because I found a film script in Simplified Chinese online! Using a ready script saved me lots of time on the TEP step (“TEP” stands for translating, editing, and post-editing) since I wouldn’t allow machine translation to ruin one of my favorite movies.
Dubbing
I recorded the vocal parts via my phone and edited them on Adobe Audition. The most challenging thing of the dubbing step was differentiating between Mirabel and Isabela’s voices.
In Audition, I went to Effects – Time and Pitch – Stress and Pitch (process)… to change pitch of the vocals so I could make the voices match the characters’ voices in the source video. Once I adjusted the voices to what I was satisfied about, I exported the .wav files and imported them to Adobe Premiere Pro.
Subtitling
After dubbing the video, I started creating subtitles in Simplified Chinese. A powerful tool I used was Adobe Premiere Pro. Premiere could transcribe the speech in the video to text in order to text files for closed-caption subtitles. This feature allowed me to transcribe the vocals in the target content and for me. You can tell that the scripts were not organized well, corresponding to the timeline.
Don’t worry. There was another cool feature Premiere had! It was creating subtitles automatically based on the text that Premiere had just recognized following Netflix Subtitling Standards.
Just one click and everything was done. Wasn’t that amazing? After, all I needed to do was post-editing the subtitles manually.
Final Delivery
Here are the source video and the localized video. Please enjoy them 😀
Takeaways
My biggest takeaway from working on this project is that I got a chance to use skills I learned from my past audio localization experiences in practice. Also, I treasured the opportunity to try out some new tools I hadn’t learned or heard about before. For example, I used Moises to remove vocals from the source video. What surprised me while I was exploring this tool was how many features Moises supported. Besides separating vocals, drums, and other instruments in any audio files in one click, an audio speed changer, and pitch changer were also available on Moises. Overall, I am quite proud of my final outcome, and it was an excellent opportunity for me to review and practice my audio-visual localization skills.
Work Cited
https://www.youtube.com/watch?v=UJig51VfoZY