4

I occasionally run into situations where an assisted listening system (commonly used for hearing aids and unobtrusive live translation at events and also for tour groups) but do not want to invest in dedicated hardware. We wouldn't use it enough to justify the cost even though when it does come up it would be very convenient to have. Several times in the past few weeks I did simultaneous translation by whispering to a group of ~10 in a corner of the same room as the speaker (seats up to 100, all acoustic, no amplification for the speaker). This is difficult because I have to speak loud enough to be heard in my corner but quiet enough not to be disruptive (and so I can hear what I'm going to translate!). Ideally these people would instead integrate with the others and just listen to their translation feed using earphones and I as a translator could be the only one needing to be unobtrusive.

My use case is for live translation, so many of the issues addressed by specialty hardware are not even issues for us. Stream latency for example is not really an issue. Being compatible with hearing aids is not an issue.

I was thinking about it last night and realized in most cases I all the hardware tech needed is already in the room. Nearly everybody has a smart-phone or tablet and a pair of earphones. I, likewise, have a phone and a microphone for it. All of these devices are powerful enough for near-real time audio input and output as well as wireless networking. It occurred to me: why couldn't this be done with an app?

Is there an existing app solution out there that:

  • Allows one mobile device to record and broadcast an audio stream over a local network.
  • Allows other mobile devices to "tune in" and listen to this audio stream.
  • Supports at least Android and iOS devices with a minimum amount of setup. This would need to be something installable from app stores without rooted devices. Instructions for installing would be given before the start of an event, but it shouldn't be an ordeal to get everybody connected.
  • Operates in some sort of multi-cast mode so the WiFi doesn't saturate with separate streams for every listener (20+ clients on the same network shouldn't cause it to wig out).
  • Operates without an off-site server so that our sometimes flaky Internet wouldn't be a hold up. An on-site media server/relay of some kind would be okay as long at it could run on Linux. Completely peer-to-peer with mobile devices would be preferable.
  • Some expense for a license to use the app in broadcast mode would be acceptable (the alternative is hundreds of dollars worth of specialty hardware) but the client app or mode should be free to install and run.

Ustream came to mind but the dependency on an Internet connection, separate data streams for each client and the focus on video angle make it a clunky adaption. Setting up a Asterisk on a local box and having everybody use a SIP client to call into a conference room that had the translator's call be the only one off-mute also occurred to me. But is there such an app out there ready to roll? If not, what components might be easiest to hijack, combine and press into service?

  • Bonus points for being able to operate ad-hoc without an access point, basically a broadcast walky-talky mode.
  • Exta bonus points for the server component being able to also act as a client simultaneously to broadcasting and for being low latency so the translator and others could use a second channel as a more standard assisted hearing device for the main speaker. Definitely not a requirement.
1
+50

Give Mumble a try.

It allows any number of clients to connect to a server and create/join channels, you can set permissions so that any client can't just create new channels or even create some listen-only channels where new clients need permission from the administrator to talk.

As for the server, it runs on Linux and it's a lot easier to set up than Asterisk, you can execute a client on the same machine thus fulfilling your requirement of the server also being a client (technically they are still two different processes, but on the local machine the latency will be lower than 1ms so it's close enough).

For mobile apps they even have an iOS app so I assume they also have the Android app, and it's very easy to use : type in an IP:Port (you can save it) and an optional password, tap the channel you want to join and that's it. Optionally you can configure push-to-talk if the auto voice activation isn't good enough.

  • 1
    This sounds promising, but your answers seems to be focused on the desktop client, but all my usage would be mobile (including the server/client combo I mentioned). Have you actually used the Android / iOS clients? Whether they are suitable for this task would really be the make or break thing here. Could they operate ad-hock or is the server side component mandatory? – Caleb Dec 20 '14 at 14:18
  • 1
    @Caleb I've used the iOS client with success; the server side is mandatory though but it's really lightweight. – user111 Dec 20 '14 at 15:24
0

This is exactly where assistive listening technology is headed- Listening on your own device via multicast wifi! I would highly suggest you check out Listen Technologies and there "ListenWiFi" system. It uses a server and a WAP to broadcast to smartphones, up to 60 users per WAP, and only requires a Linux server and audio interface (ADC). Now I know you want to avoid hardware altogether, but this is as close to what you speak of that I know when it comes to assistive listening. The good news is its not near as much as a traditional ALS system, the apps for iOS and android are free, and people can simply bring their own device. No Internet required, no license required, just plug and play! Check out the new system here: http://www.listentech.com/assistive-listening/listenwifi/

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.