I am training to be a “Simultaneous Interpreter” which is a misnomer; it really means that the interpreter must simultaneously listen to a foreign speech and produce an English equivalent speech with a lag due to the necessity to first understand the meaning. (“Interpreting” means converting a foreign speech into English speech; “Translation” means only written communication.)
What I am seeking is a setup to a) Live record of a foreign speaker, b) Playback with a controllable lag or delay of 1-20 second and c) recording on another channel of the interpreter’s English speech.
The “lag” between the foreign speech and its playback is to be variable, controlled by a “hold in buffer” feature; this would allow the interpreter to take the time needed to convert the foreign speech into English. The point of “Simultaneous Interpreting” is not to do instant dubbing into English, but to produce and accurate a natural sounding English rendition - whether if it takes seconds to minutes of lag to produce it.
Of course, there should be an “instant replay” when one can repeat a sentence from the foreign speaker.
We can’t do “B.” The live delay thing has been requested several times and we still can’t do that.
When I had to make a delay, I ended up with a beginner package of a much larger, expensive sound program which would do live effects and sound processing. “Beginner” worked perfectly for me because I only needed the delay.
So that’s your Google search term.
I have a friend who does Spanish/English simul-translation for the court. When coming back to English, an inexperienced judge always asks what’s taking so long.
“Because Spanish doesn’t get to the point until the end.”
The judge, like most people who only speak one language, doesn’t appreciate that you must hear the full context to understand the meaning intended. It takes time to process and come up with a natural sounding English version.
Could one just halt playback after a sentence or paragraph manually?
Let’s try this. You have a stand-alone sound player with the foreign language on it. It plays into one side of the Audacity stereo recording and the translator’s headphones. The translator voice goes in the other side. Audacity starts recording and stays constantly recording through the whole performance. Start the foreign monologue and pause it while the translator works. Take the player out of pause for the next segment. Repeat.
Audacity will produce a ping-pong effect with foreign, say, on the left and the translation on the right. Split the show into two mono tracks to apply effects, cut and clean up where necessary. When you export to a sound file, Audacity will produce a mixed show with both voices in the middle; one mono track. It should sound like a real trial.
I don’t know any way to get that effect by starting and stopping Audacity. You’re close to what happens in musical overdubbing where you play all the instruments one after the other and layer them into a finished song. I can’t see any way of getting past the first foreign monologue, because you can’t pause Audacity and record the translation at the same time.
So you would start by listening to the person in “real time” (as they speak).
You then speak the interpreted version.
I presume that you want the delay so that you have time to speak the translation and “pause” the foreign speaker while you do so?
So you pause the foreign speaker for a couple of seconds and speak your interpretation.
The foreign speaker recording may now be resumed, but it is now delayed by a couple of seconds.
How would you “catch up” with the foreign speaker, or does your interpretation become increasingly delayed?
“Audio to buffer” in Audacity’s Recording Preferences delays playback by the amount you enter. It does not delay playthrough of a recording or delay recording itself (at least, not on Windows).
Whether you can use those as part of your solution, I don’t know. Have you searched for specialist commercial translating software?