Mihael wrote:Nick, one question, i am just curious, how did you made Pops Mensa Bonsu speech id?
Mensah-Bonsu in particular was included in NBA Live 08. So I just exported him from there.
In general, Packs I to III contain players from Live 07 and 08 speech databases. Since IV, they are entirely my creations. This has a few exceptions here and there, but as a general rule, it applies.
For example, many of the really excited speeches from I to III were added by me. Also, if my version was, in my opinion, better (especially when I have first and last names available in the database - eg. Aaron Gray), I kept it, as Al Murdoch changed a little bit his voice and his style throughout the years (and I tend to prefer his early work, from 97 to 2003 or 2004). It's a slight difference, but noticeable when working with sound editors.
Then, my later packs contain a couple of Live players (Roy Hibbert, for whatever reason, has a SpeechID in Live 08). Some of them are very tricky, like Mbah a Moute, who contains segments from 3 or 4 different players, and I've tried so many times to get it right that I forget the source. Krystkowiak, for instance, comes from Larry (Bird, Costello, I don't remember), Doug Christie ("Kryst"), Damir Markota ("ko"), and Wally Szczerbiak ("wiak"). It sounds like "biak" rather than "wiak", but in-game, it works nice.
Sometimes it's difficult to come up with something decent, but I try my best. Besides, intonation is fundamental. So, first names can only be covered by first names, and last names by last names, because of how Murdoch gives a rising intonation after first names. Moreover, stress is also decisive, as it influences in how a given work is pronounced. For example, "Marcus" and "Martell" contain the same first syllable, but the stress varies, so it's not the same. If you want to create "Marty", Marcus is your choice. On the other hand, if you want to create "Marquis", Martell works best.
This is a small explanation regarding the speech creating process. Of course, there are other factors involved (like the different tones used in the really excited lines, which on occasions don't allow to mesh them together, unless they sound terribly awful -a syllable very low-pitched followed by a high-pitched syllable, for example; or sometimes the syllables cannot get together because you are forcing them to, generating some clipping or a distortion). But I think you get the point without me getting too technical.