|
|||||||||
|
Thread Tools | Search this Thread |
October 9th, 2005, 05:16 AM | #1 |
Inner Circle
Join Date: Jan 2004
Location: Boca Raton, FL
Posts: 3,014
|
How would you record a language translator?
I have an interview production where the questions and answers must be translated to another language. I read a book by Jay Rose that gave some tips but the logistics of the translator needing to listen then translate and maybe rewind to listen again before recording are in need of some technique. I can come up with something but thought I'd ask if anyone had come up with a technique,
|
October 9th, 2005, 08:03 AM | #2 | |
Inner Circle
Join Date: Mar 2005
Location: Hamilton, Ontario, Canada
Posts: 5,742
|
Quote:
Are you going to try to do a lipsync'd translation or VO? You'll be doing yourself and your translator a huge favour by providing him or her with a verbatim transcription of the source audio well in advance of the recording session. And I mean verbatim - if the source dialog goes "..in 1932 ... or was it '34 ... I was working for ummm United Widgets" that's exactly how it should read on the transcript, with all pauses, digressions and restarts, grammar mistakes, "um's" and "ah's" exactly as they appear in the recorded track. Best use a workstation that has a full-duplex sound card that lets you simultaneously record on some channels while playing back others. A multichannel audio interface with ASIO drivers intended for multitrack recording allows you to do this but run-of-the mill-sound cards won't. This allows you to view/listen to one set of tracks while recording new audio tracks in parallel. Most NLEs and audio editors allow you to loop a segment. Some allow you to record "takes" where it loops during recording and each pass through the loop stacks a new take on top of the old without deleting the old version. Then after the session you can choose which take to use in the final. Some, like Nuendo, make it easy not only to pick the best take but even to slice up all the takes and assemble the best segments from each one into the final version. Set up for recording where your translator can see a video monitor and listen to both the source and themselves through headphones, preferably where they can adjust the level of each signal in the phones themselves to their liking. Put your video and source audio on playback looping over the area you're working on, create a fresh, parallel, audio track for the translator, and when the talent is ready, arm it for recording and go for a take. Be prepared, it can be a very laborious process. |
|
October 9th, 2005, 01:14 PM | #3 |
Inner Circle
Join Date: Jan 2004
Location: Boca Raton, FL
Posts: 3,014
|
I am not lip sync'ing. I see the concept of having a setup and rig that allows multiple takes and organizes them. Thanks.
|
October 17th, 2005, 05:49 PM | #4 |
Inner Circle
Join Date: Jan 2004
Location: Boca Raton, FL
Posts: 3,014
|
This is just a followup to say that the session went well. Having the translation written down was hands down the best tip and simplifies things tremendously. After doing a couple recordings, the translator was able to don headphones and listen to the english at a low volume and key off the interviewer and interviewee voices to know when to say the corresponding translated text. There was no translation on the fly, just self cueing and reading. Of course, creating a transcript of the interview was no small feat. I had a fast typist and it took 3 times the length of the interview to type it in. However, this saved studio time as it made the recording basically real time plus setup. Thanks again.
|
| ||||||
|
|