Typically, without using trickery, magic, or violence, a creature can't make another creature absorb a pearl of speech (Magic Item Compendium 118) (600 gp; 0 lbs.) as its description says
When you place a pearl of speech upon your tongue (a standard action), it is absorbed into your mouth until you speak the proper command word to release it.
(Emphasis mine.) So an owlbear (Monster Manual 206) must itself—for whatever reason—place the pearl of speech upon its own tongue for the pearl to be absorbed. This isn't an impossible thing to convince an owlbear to do, but doing so usually entails some risk. For instance, a low-level ranger probably must be fairly optimized for the wild empathy special ability to get an owlbear to do this.
That said, once the owlbear does absorb the pearl, the owlbear can then communicate in the language the pearl provides… to the best of its ability, anyway. That is, this reader suspects that the owlbear's physiology sadly still limits it to mostly hooting and growling. See, while the pearl "grants [the creature] the ability to speak and understand a specific language," it doesn't also change the creature's underlying (albeit admittedly magical) biology. The pearl's description says that pearls were "created by drow and used to command their slaves without stooping to learn their foul languages," not that they were, for instance, created by druids to give animals the gift of speech!
Thus this reader believes that, as written, an absorbed pearl essentially grants the creature 1 rank in skill Speak Language as if it had picked Common, Elven, Slaad, Clockwork Horror, or whatever language the pearl contains, but that virtual skill rank doesn't, in turn, permit an owlbear—like a disturbing combination of Yogi Bear and Woodsy Owl—to actually and for-reals talk.
However, this player and this DM knows that's no fun. Like, at all.
Thus, in this DM's campaigns, when a normally noncommunicative creature absorbs a pearl of speech, that creature gains the full capacity to communicate in the language the pearl provides out loud and for-reals using actual speech, just as if every listener were affected by the 1st-level druid spell speak with animals (Player's Handbook 281). Further, like that spell, the pearl's presence "doesn't make [a creature] any more friendly or cooperative than normal[, and] wary and cunning [creatures] are likely to be terse and evasive, while the more stupid ones make inane comments." So, in this DM's campaigns, that owlbear will be making a lot of inane comments. All the time. And this DM says, sincerely, thank you for paying 600 gp to give me that role-playing opportunity.
Similarly, this player would try to convince a DM who was reluctant to rule likewise that the pearl should work this way in that DM's campaigns, too, if for no other reason than because the pearl is a lot of fun for the DM.
Note: In this DM's campaigns, pearls of speech are frequently used by druids, paladins, wizards and others so that their animal companions, familiars, and special mounts can talk. My campaigns have only improved as a result.