audio jerry-rig advice

[jd-science.livejournal.com]

i am helping someone with a multi-party (25 person) video-over-IP workaround and need some audio advice. Here are the conditions:

1. Party 1 needs to be able to see and hear Parties 2-25.
2. Parties 2-25 need to be able to see and hear Party 1.
3. Parties 2-25 don't need to see each other, but do need to hear each other.

We are using Sightspeed, which can host up to 9 people on a single video call, allowing Party 1 to see 8 others at a time. We plan to have Party 1 on three separate calls/computers, to satisfy conditions 1 and 2. Within each call, everyone can see and hear each other, but they cannot hear/see across calls (e.g., Parties 2-9 can see and hear each other, and Parties 10-17 can see and hear each other, but Party 2 cannot see or hear Party 10). The lack of cross-call video is fine, but the lack of audio crossover violates condition 3.

So what I'd like to do is find the least painful solution that will minimize feedback. What I'm thinking is that we need four mixing boards, but I also believe that there is probably a more streamlined solution. I'd also like to know if what I'm proposing is impossible. I am very inexperienced with audio wizardry and would welcome any feedback (ha ha!).

thanks.

Comments

[jd-science.livejournal.com]

oh yeah - sorry if this is not DS-related enough. I see it as being so, given the density of people in the neighborhood who probably enjoy answering questions like this, but i'm happy to remove the post if it's inappropriate.

[ron_newman]

It's fine to post here, but I haven't a clue how to tag it.

[thebinturong.livejournal.com]

It's essentially a two party conference with one party being alone and the other together in a virtual room. And the video and audio can be handled differently. The video can be your IP video capture. And if you have access to Plain Old Telephone Service, then you can consider teleconferencing to handle the audio. If anyone is in the same room, you can get a teleconferencing device for that room. That will probably have some intelligent muting. If everyone is just calling in on their own, everybody hears eachother and can mute their microphone and speaker on their phone as needed.

[thetathx1138.livejournal.com]

It shouldn't be IMPOSSIBLE, but you are going to need several mixing boards and it's going to have to be VERY organized to keep from peaking the main mix.

[jimmyfergus.livejournal.com]

Sorry to be the unhelpful pedant, but it's jury rigged, or jerry built, but not jerry rigged.

I couldn't help myself.

[jd-science.livejournal.com]

i actually looked this up because i had heard that before, but decided to flout tradition/standards and go with the way i say it, which is "jerry-rigged." :)

[jd-science.livejournal.com]

do you think 4 is enough? what sort of pitfalls should i be looking for? thanks.

[jd-science.livejournal.com]

The dual-medium format is how we did it in the trial run - everyone complained about having to coordinate both the teleconference and the videoconference. They wanted a single-application solution.

I hadn't thought of it being a two-party conference, one alone the other in a virtual room. I wonder if/how that helps me... hmm....

[thetathx1138.livejournal.com]

Depending on your inputs, should be. Your problem is the main mix. As long as people take turns speaking and don't try to talk over each other, you should be OK.

[thebinturong.livejournal.com]

If the problem with the phone was the difficulty of too many parties coordinating mutes, I think that adding audio operators who are isolated from conferencers is just adding to the complexity (my 2c) If participants can use headsets or handsets, that will help alot with feedback. I think it would be a necessity for party 1 to use a headset. Otherwise the bottleneck seems to be the software. Unfortunately I'm not familiar with IP video conferencing systems so I can't be helpful with that. Best of luck!