Thread for feedback on changes.
banner art by Michal Lisowski
And mechanically it'd make sense that the voicenote would pick up that shout. I get the reasons why broad room recording isn't a thing right now, but that's my feedback on shout and the VoiceNote.
6.A plus introducing falsification tools is a ticking time bomb. Manufactured allegations against virtual persons in virtual spaces can and will impact players in the real world. Just claims alone can destroy players and spread outside the game, just imagine what perfectly simulated logs that can be corroborated by multiple players will do.
You can play the VoiceNote's recordings back to the room when held or on the ground, but it doesn't play when hidden.
I think there's some of us who think it should also playback while hidden, whether in the room or behind an object.
There is already evidence to show that false accusations against characters have bled through the veil and become accusations against players. I agree it's probably something for a thread of it's own but I think there is a myopia that this kind of stuff is just going to be fun gameplay, when it's actually a dangerous introduction without strict policies to control it, something we don't have currently.
Otherwise, the question is how do I recognize a fake voice vs a real one?
My proposal in the short-term would just be any forensics check against a PC's voice not check if they're disguised but rather challenge against their disguise skill defacto. It's not ideal because then you can just extrapolate well-disguised people by being indiscernible but non-contested checks kinda ruin the point of PvP.
I love the idea, but I am curious if the implementation has considered many of the aspects of a voice and language.
You grab a recording from an e-memory. You plug it into the VOCAL device. Then, you record a sample from someone you want to compare that voice from. You pick what position of the message you are comparing that voice from (since a recording can have more than 1 voice / message).
Based on that, the device tries to study behavior patterns to create a similar wavelength based on speech patterns or what ever have you to determine how likely it is that the two voices come from the same source. In modern times, what that device uses is neural networks and AI to do the same.
Does that make sense?
@Reefer
What if a master disguiser says something in a recording and then they want to verify it is them? It wouldn't make too much sense to be unable to prove the voice as yours right? I was about to implement it like that, but then this kind of questions made me stop if that makes sense.
Is this an oversight or is the intent to allow false negatives and positives regardless of skill?
You do not need to answer the question, as I think it verges on FOIC, just something people should be aware of across the board perhaps?
There is the potential in a LOT of code new and old, for OOC abuse. Don't do it. We have a Community Harm rule that allows the staff to act on things that are harming the community.
In terms of IC: If someone has a decent disguise skill, you can probably make your voice sound somewhat like someone else's. This has always been the case, and people use it for things like Progias and Radios already. That's why it is important ICly (and OOCly) to recognize that the person you think you are talking to, or the person you think you are hear, may not actually be that person.
Everyone needs to approach these systems with a sense of maturity and rationality. Just because you can OOCly do something (like set your @voice to whatever you want) doesn't mean your character can ICly do that. Take these things into account and feel free to xhelp if you aren't sure if your character would be able to do something, and you're looking for feedback on it.
If you record an hour-long conversation, will it take an entire hour to listen to the playback?
Could there be a button to switch the playback speed to 2x or 4x the original speed?
Could you connect the recorder to an e-print to print out a transcript of a recording? (Or print the .mp7 file from an e-memory module and an e-note?)
Let's say you're recording an interview and you ask someone a question, but they take a good 5 minutes to answer. Will you have to wait through 5 minutes of silence on the playback before it plays the person's answer? I wonder if it could be possible to have the playback on a timer sort of like a NLM TV show, outputting dialogue every 10 or 15 seconds or so (dependent on what speed you set it to).
As it stands, you would need to wait it. The reason for this is to preserve the pause a person gives when speaking, and give people that record anything in their recorders the creative right to decide on the pacing.
In theory, we could code something for speed to be 2x or 4x, yeah.
In theory, the e-print transcript suggestion is possible too.
Those two are possible solutions!
I think preserving the speed the talker talks at gives more creative freedom than making everything arbitrarily 10 or 15 seconds, but what does everyone else think?
I like the idea of a speed up option, but maybe it requires a copy of the file to do that rather than altering/speeding up the original recording/file.
Personally, I would feel anxious if I knew that every moment I took to compose what I want to say while avoiding typos, meant that anyone listening to the file in perpetuity would have to wait 1-2 uncomfortable minutes for my response. Some conversations go at a fast pace but unless you have what you want to say already prepared - which, in theory, I suppose you could 'rehearse' the conversation/interview/monologue in advance - then it may take some time to type up a quality thing to say.
Speaking of this translation feature, I've noticed that it doesn't indicate the original language someone was speaking in. Not sure if that's intended, or if it's more an OOC function of "omitting language altogether is less intensive in terms of file storage."
There are ways to not include the accent for sure. However, it would make it hard to understand when there is a shift of speaker. I am eager to hear more opinions on this transcription idea and what everyone thinks. I also like the idea of the speed up.
@svetlana
NLM TV has the option for people to record at their own pace and I believe very few of the current shows use the default pacing of 10-15 seconds.
I think the option to edit these recordings in the future to maybe modify the timing would be interesting, although that would be a big feature.
I am also curious to hear more thoughts on this dilemma.
@Amiga
The 'translation feature' was decided upon after discussion mainly to avoid the intricacies of language & file storage. We also came upon the realization that in 2023 most devices have a translation feature that works like a charm, so in the 2100s these devices would likely be highly advanced and have this functionality. The same intricacies and issues are what make TVs 'translate' stuff for you too.
I love the opinions on recorders & VOCAL.
On the same note, are people using these / enjoying these? Keep the feedback coming if any more!
I'm sure I'm missing others that are visible in the code and would be welcome additions.
Lastly, and mind you, this is a pie in the sky goal but Cyber-Design and Biomods need to be aligned. They should be using the same variables and dare I say offer more extreme changes such as complete @skincolor changes and the like. I have some past idea threads about stuff like this that may be worth bumping if their is interest.
I think making an option to uncache/erase the currently cached the file from the VoiceNote would be helpful. This would create the option to keep the cached filed on there if you want, but also erase and start fresh.