1

What is the best way to go about recording custom greetings?

*This may sound sad, but I am trying to wish as many of my friends a happy new year, but do not want to spend all the time in the world calling each of them.

How would you suggest I go about this project?

My thoughts were as follows:

  • Record Individual Names

    • I was thinking about trying to space it out so I can programatically, cut the audio so that I could trim it to sound natural, meaning I would continuously record :
      • " Happy New year Jim ...... Happy new year you crazy animal, ........ happy new year Bob, this is your year!!.........."

  • Or is it better to try and record each persons name one by one and then have a few pre-recorded audio files and then stitch them together ( loop through them)?

The only issue I see here is how do I make it smooth so that it does not sound like it is chopped up

  • Any suggestions on how to automate some of this, because I know audio editing can be incredibly time consuming. ( Do not get me wrong, I am willing to put the time in)

    • I am currently running Linux, but have the ability to get to a Macintosh or Windows Machine.
    • I have a Blue Snowball microphone ( Blue Snowball iCE Condenser Microphone, Cardioid - Black)
Improve this question
4
  • 1
    Hi Ted - we don't do recommendations, so I have deleted that part of the question. You can use any software you like (and scripting is probably easier for you on Linux)
    – Rory Alsop ♦
    Dec 29, 2016 at 14:24
  • @Rory Sorry, i did not know, thank you for editing!
    – Ted Taylor of Life
    Dec 29, 2016 at 14:26
  • No worries Ted - there is a Software Recommendations stack exchange site, and scripting is on topic on SuperUser so if you don't get anything that works here, we can look at migrating it over to one of them as appropriate.
    – Rory Alsop ♦
    Dec 29, 2016 at 14:35
  • I really appreciate it. People like you are part of the reason I do everything in my power to help those who ask for it. Have a great day!
    – Ted Taylor of Life
    Dec 29, 2016 at 14:51

2 Answers 2

Reset to default
1

I'm going to throw an answer out there as the audio part of this question falls within domain. The scripting side is debatable but probably not worth moving to another SE as it's a trivial answer.

The audio software you will need to use is SoX (Sound Exchange) and FFMPEG. You will script in bash and/or python.

The methodology will be as you have suggested - record names in a loop, however I would counsel you to take care to search for suitable edit points. This might mean you have to append words prior to the names and record the names multiple times depending on the context. Some words that you will use might "run in" to other names depending on how they start. If you get this right, it will sound much more natural.

Improve this answer
2
  • Thank you for your answer! "however I would counsel you to take care to search for suitable edit points." Are you suggesting that I always say a name and cut off at a certain point meaning that no matter how many times I say "Hello x" I cut off at say y seconds, always? That is a great idea!
    – Ted Taylor of Life
    Dec 31, 2016 at 14:04
  • 1
    Look for transients within names and words - Transients make the best edit points as the sound of the edit can be concealed within the transient. For instance "Peter" has a transient at the start of the "P..", as does Tom, Thomas, Tammy etc. However You will need to take care with names such as "Emily" as the words of "Hello Emily" will run together with no noticeable transient that can be used as a good edit point.
    – Mark
    Jan 1, 2017 at 0:05
1

Perhaps not the answer you really wanted, however...

I think your life would be a whole lot simpler if you just made each one in its entirety. Recording them all in a line, reading from a script would be fine - you could chop them up afterwards in pretty much any audio recording package.
Audacity is free, but it looks like you'd have to build your own on nix.

I think the potential for sounding like an automaton, or a badly-edited railway announcement is just far too high if you chose to record the greetings & the names separately & then try to splice them together.
I also think the editing process itself would be more work than it's worth, even if you got past the railway announcement issues.

Improve this answer
1
  • Thank you very much for your response. I will put some thought it what you said. You may be very well right. My only concerns are the scaling factor of this method.
    – Ted Taylor of Life
    Dec 31, 2016 at 14:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.