James Keith McElveen — Direct/Cross
152 linesJUDGE EVERETT: Please call your next witness.
MS. CAPPLEMAN: The state calls Keith McElveen.
JUDGE EVERETT: Please raise your right hand, sir.
JUDGE EVERETT: Do you swear or affirm the testimony you're about to give will be the truth?
JAMES KEITH MCELVEEN: I do.
JUDGE EVERETT: You may take your seat.
MS. CAPPLEMAN: Sir, please introduce yourself and spell your name.
JAMES KEITH MCELVEEN: James Keith McElveen. J-A-M-E-S, K-E-I-T-H, M-C-E-L-V-E-E-N.
MS. CAPPLEMAN: How are you employed, sir?
JAMES KEITH MCELVEEN: I work as a forensics engineer — a media forensics engineer.
MS. CAPPLEMAN: And what is a forensic engineer?
JAMES KEITH MCELVEEN: We specialize in taking audio, video, or computer data and extracting intelligence out of it.
MS. CAPPLEMAN: How long have you been doing this type of work?
JAMES KEITH MCELVEEN: A bit over 35 years.
MS. CAPPLEMAN: Can you go through your training and experience that qualifies you to do this type of forensic engineering?
JAMES KEITH MCELVEEN: I began doing this type of forensic engineering about 35 years ago, working for the government, working on war crimes investigations, unmixing voices.
JAMES KEITH MCELVEEN: Since then, I've had 20-plus, either given or taken, 20-plus courses in this discipline.
MS. CAPPLEMAN: All right. And have you had an opportunity to testify as an expert in your field previously?
JAMES KEITH MCELVEEN: Yes, I have.
MS. CAPPLEMAN: Judge, I would tender the witness at this time as an expert in the area of forensic engineering.
JUDGE EVERETT: Do you wish to engage in any further voir dire of the witness?
MR. RASHBAUM: No, Your Honor.
JUDGE EVERETT: Is there any challenge to his ability to provide an opinion in the stated field?
MR. RASHBAUM: No.
JUDGE EVERETT: Members of the jury, the witness will be permitted to provide you with an opinion today in the field of forensic engineering. You may continue.
MS. CAPPLEMAN: Sir, were you asked to clarify recordings in reference to the case State of Florida versus Charlie Adelson?
JAMES KEITH MCELVEEN: Yes, I was.
MS. CAPPLEMAN: All right. I want to start with the Dolce Vita restaurant recording. What items were provided to you in reference to that exhibit?
JAMES KEITH MCELVEEN: I received two recordings — one from each device that was deployed in that restaurant.
MS. CAPPLEMAN: All right. And was one better audio and one better video?
MS. CAPPLEMAN: And what's the part that's of interest to you?
JAMES KEITH MCELVEEN: The audio.
MS. CAPPLEMAN: Okay. And so you don't do anything with the video?
JAMES KEITH MCELVEEN: Only used it for observational timing reasons.
MS. CAPPLEMAN: Okay. And was the bag that had the mesh on it — was that the one that was the better audio, or the one that had the better video?
JAMES KEITH MCELVEEN: Right. The one with the better video had very poor audio. The one that had better audio had very poor video.
MS. CAPPLEMAN: All right, I'm with you. And the one where we're looking through the little holes — is that the one with the good audio or the good video?
JAMES KEITH MCELVEEN: The good audio.
MS. CAPPLEMAN: All right. Can you sort of walk through the process for the jury of how you clarify an audio that has a lot of background noise going on?
JAMES KEITH MCELVEEN: Yes. Our technique is a bit different than other techniques in that we take the audio and treat it like it's an image — like a photograph that's out of focus, it's blurry. And so we read it into the computer and figure out how to unblur the sound, just like you would unblur a photograph. And this fortunately has the nice effect of focusing in on the person's voice, or the people's voice if they're near each other, and blurring out the rest of the background — sort of how you would use a video lens if you're a photographer and you can zoom in on a person's face, but everything behind the person and everything closer is blurry and kind of pushed down and not as noticeable.
MS. CAPPLEMAN: Were there any particular issues or problems with this audio?
JAMES KEITH MCELVEEN: It was, frankly, a mess.
JAMES KEITH MCELVEEN: Competing voices, loud noises, movement of both devices, not good positioning. There was a kitchen.
JAMES KEITH MCELVEEN: And of course it's a busy cafe — coffee grinders, blenders, order taking, lots of different sounds going on.
MS. CAPPLEMAN: All right. And can you tell the jury that the most up-to-date efforts available in your field were applied to this exhibit to make it as good as possible?
JAMES KEITH MCELVEEN: They were.
MS. CAPPLEMAN: All right. And what was the outcome? I think you've got some different segments — could you tell us about that?
JAMES KEITH MCELVEEN: Yes. Fortunately, the recording device that had the better audio was also the one that had the longest audio. The recording started at different times — one device arrived into the establishment significantly prior to the other one. I believe it was like 20 minutes' difference.
JAMES KEITH MCELVEEN: The one that started first — its recording, the person carrying it came in from the outside, came to the counter, then went to a seat, and then moved to a different seat, getting better position.
JAMES KEITH MCELVEEN: And each one of those changes of location causes the focus to change — just like if you were taking a picture and all of a sudden the person you're trying to take a picture of moves, you've lost your focus, and you have to regain your focus.
JAMES KEITH MCELVEEN: Also, a limitation of our tool limited the sheer amount of audio. It's a huge amount of recording, and so we had to break it up into portions, segments, which we did — the beginning, the middle, and the end — and we worked with the FBI agent to break it at about the same points that the motion happened of the recording device.
MS. CAPPLEMAN: Okay. And have you had an opportunity to examine State's Exhibit 110 — thumb drive?
JAMES KEITH MCELVEEN: Yes — immediately before we came in here.
MS. CAPPLEMAN: All right. And is this a fair and accurate work product that you did, as far as the audio only on the Dolce Vita recording?
MS. CAPPLEMAN: Okay. Judge, I'd ask to move into evidence State's 110.
JUDGE EVERETT: Any objection?
MR. RASHBAUM: Okay. Subject to previous. Objection.
JUDGE EVERETT: State's 110 will be admitted over objection at this time.
MS. CAPPLEMAN: Is there any part of your processing or methodology that actually changes the voices or adds words into the voices? Anything like that?
JAMES KEITH MCELVEEN: No. When we begin our processing, we set up the controls, and the tool does not allow you to actually make any changes.
JAMES KEITH MCELVEEN: You hit "process" and it starts at the beginning and works all the way to the end, and you can't change anything as you go along and decide to leave the word "no" or "yes" out.
JAMES KEITH MCELVEEN: You get the entire thing, and the same length of the recording in is the same that comes out — down to microseconds.
MS. CAPPLEMAN: And did you also perform a similar analysis or work on a second restaurant recording, known as the Matsuri restaurant recording?
JAMES KEITH MCELVEEN: I did.
MS. CAPPLEMAN: And have you had an opportunity to review State's 113?
JAMES KEITH MCELVEEN: Yes. I mean, before — I'm sorry, go ahead.
MS. CAPPLEMAN: Immediately before?
JAMES KEITH MCELVEEN: All right — before we came into the courtroom.
MS. CAPPLEMAN: And is this the thumb drive that you prepared with your work in reference to the Matsuri recording?
JAMES KEITH MCELVEEN: It is.
MS. CAPPLEMAN: All right. And were the same types of processes applied to this recording as the first one we discussed?
JAMES KEITH MCELVEEN: Exactly the same way.
MS. CAPPLEMAN: Moving into evidence, State's 113.
JUDGE EVERETT: Any objection to State's 113?
MR. RASHBAUM: No.
JUDGE EVERETT: State's 113 is admitted at this time.
MS. CAPPLEMAN: No further questions, Your Honor.
JUDGE EVERETT: Cross-examination.
MR. RASHBAUM: Just briefly, Your Honor.
MR. RASHBAUM: Starting with the, uh, Dolce Vita recording. That recording was made in April of 2016, right?
JAMES KEITH MCELVEEN: I do not know.
MR. RASHBAUM: You were asked to enhance the recording in January of 2022, right?
JAMES KEITH MCELVEEN: That sounds correct.
MR. RASHBAUM: You'd agree with me that there's 20-plus minutes of the recording that's missing, right? The first 20 minutes or so of the recording are missing?
JAMES KEITH MCELVEEN: What do you— please clarify what you mean by "missing."
MR. RASHBAUM: Well, fair point. You would agree that the meeting that was recorded started before the recording, right?
JAMES KEITH MCELVEEN: If you mean how long the conversational participants had been carrying on the conversation before the person carrying the recorder entered the restaurant, I have no idea how long they had been talking.
MR. RASHBAUM: The recording picks up in the middle of a conversation, correct?
JAMES KEITH MCELVEEN: Correct? If you mean device two, the one with the longer recording, I believe that you might be confusing the fact that it was the middle segment of three, if you mean the conversation. I think you mean—
MR. RASHBAUM: The recording that you enhanced is not the beginning of the conversation, correct, between the participants?
JAMES KEITH MCELVEEN: Again, I don't know when they began talking, but we cleaned up all the way from the beginning of the recorder entering the restaurant — by the front, to a chair, to another seat. And the first segment had clear conversation going on at times; voice level goes up and down.
JAMES KEITH MCELVEEN: Then there's the middle segment, which is — a few seconds, 30 seconds or something, maybe a minute, go by — and then the video from the other device starts.
MR. RASHBAUM: Fair enough. If that's the one that you're talking about, then that's only the middle.
MR. RASHBAUM: The lead up to that portion is another 20 minutes. But you don't know if the conversation began before the recording started, correct?
JAMES KEITH MCELVEEN: Correct. I have no way of knowing that.
MR. RASHBAUM: Now, even on your enhanced version, virtually everything said by the female participant remains inaudible or unintelligible, right?
JAMES KEITH MCELVEEN: I'm not exactly sure what you mean by "virtually," but I would say that the female participant talked very seldomly.
JAMES KEITH MCELVEEN: More of her words are audible in the first segment than in the middle segment, but she also seemed to talk less. And I can say that because you can obviously tell that she does say some words even when you can't pick them out — you can tell something is being said, not only by the male participant — his conversational behavior — but you can actually hear some vocalizations from the female. But you can't always pick out what she's saying.
MR. RASHBAUM: During a lot of the recording, no matter what you did to fix that, lens, you still couldn't make out what she was saying, correct?
JAMES KEITH MCELVEEN: Yes. She was speaking away from the male participant and speaking in a low enough level that at many times the male participant asked her to repeat herself.
MR. RASHBAUM: Nonetheless, on the recording itself, no matter what you did, a lot of times you're only hearing half of a conversation, right? One side of a conversation.
JAMES KEITH MCELVEEN: One side of a conversation. That's correct.
MR. RASHBAUM: Now, were you aware that the FBI did its own enhancement back in August of 2016?
JAMES KEITH MCELVEEN: I was told that.
MR. RASHBAUM: Did you ever listen to that enhancement? Enhancement?
JAMES KEITH MCELVEEN: Yes, I did.
MR. RASHBAUM: Are you aware that virtually everything you hear on your enhancement, you can hear on that enhancement?
JAMES KEITH MCELVEEN: I wouldn't agree with that statement.
MR. RASHBAUM: Okay. You hear different words on your enhancement versus the first enhancement?
MR. RASHBAUM: Or are they just clearer?
JAMES KEITH MCELVEEN: Much clearer.
JAMES KEITH MCELVEEN: My understanding was the reason we were given these recordings was that originally, the evidence was not suitable for playback to the jury because it required somebody to have very good listening equipment and to listen very carefully and loop it many times to be able to really pick out what the words might be.
JAMES KEITH MCELVEEN: I would say, in all fairness, that after our enhancement, you didn't have to struggle that much except on those inaudible parts.
MR. RASHBAUM: Okay, so I'm going to ask a very specific question, not about the ease of hearing it. I'm just asking, could you hear any new words on your recording versus the original recording?
JAMES KEITH MCELVEEN: I can only speak for myself, but I would say yes for me, for my ears.
MR. RASHBAUM: Turning your attention to Matsuri, that recording was made in April of 2016, right?
JAMES KEITH MCELVEEN: I do not have this information, and—
MR. RASHBAUM: Again, you weren't asked to perform an enhancement until early 2022, right?
JAMES KEITH MCELVEEN: That sounds right.
MR. RASHBAUM: And for those two recordings, again, is it your testimony that you hear different words on your enhancement versus the original?
JAMES KEITH MCELVEEN: I wouldn't say... well, actually I would agree, um, different words than were originally transcribed versus what I heard, and I would say that the result of our enhancement, more words are intelligible.
MR. RASHBAUM: Now, you also note in your report that portions where one participant whispers in the ear of the other — this is Matsuri — were not recoverable due to the soft level of the sound. Is that accurate?
JAMES KEITH MCELVEEN: Yes. I would also say, and due to the obscuration of the head — one participant leans over and whispers very close to the other person's ear, so all of a sudden the head's in the way of the focusing effort we made.
MR. RASHBAUM: So in those instances where those whispers occur at Matsuri, you have no idea what was said?
JAMES KEITH MCELVEEN: Correct.
MR. RASHBAUM: May I have one moment, Your Honor?
JUDGE EVERETT: Take your time.
MR. RASHBAUM: No further questions.
JUDGE EVERETT: Redirect examination?
MS. CAPPLEMAN: No, Your Honor. No further questions.
JUDGE EVERETT: You may step down, sir. Members of the jury, before we continue with the next witness we will take a brief break. The bailiff will escort you back to the jury room.
JUDGE EVERETT: Once again, Mr. Rashbaum, the parts of the recording that have now been modified.
MR. RASHBAUM: Do we have it?
JUDGE EVERETT: Go ahead.
MR. RASHBAUM: May we be excused to do that?
JUDGE EVERETT: That's fine.
JUDGE EVERETT: Mr. Rashbaum, have you finished reviewing?
JUDGE EVERETT: I'm sorry, Jason. Have you finished reviewing at this point?
MR. RASHBAUM: We did review it, yes.
JUDGE EVERETT: Is there anything further that you wish to raise?
MR. RASHBAUM: Just our previous objection, Your Honor.
JUDGE EVERETT: Very well. As to the instruction to be given before the recording is shown, I have drafted one. Do the parties have any instructions that they wish the court to consider as well?
MS. CAPPLEMAN: I have not drafted anything, Your Honor.
MS. MEYERS: We've taken a look at pattern instruction 2.6.
JUDGE EVERETT: Okay. If you wish to come up here, you can read this. My intention would be to give this instruction along with 2.6 as well.
JUDGE EVERETT: Does the State wish to be heard concerning the instruction?
MS. CAPPLEMAN: No.
JUDGE EVERETT: Well, I will give this instruction and also 2.6 unless the parties are asking for anything else to be given as well.
JUDGE EVERETT: Yeah, please bring in the jurors.
JUDGE EVERETT: State, you may call your next witness.