Building The StreamBox
Liam Burke • June 18, 2018
From Side Project To Final Product
Like SmartSign before it, our new StreamBox started out life as a 'Friday afternoon' project to investigate a low cost and reliable way of encoding audio and sending it to web streaming hosts such as our Broadcast Player
service.
There are of course many options available but the broadly fall into two categories:
- Software Solutions
- Hardware Devices
Hardware devices provide a good alternative as they usually do not usually require any type of user interaction to do their job. The downside is that some hardware solutions can expensive.
The challenge was whether we could come up with a hardware device that would avoid the potential pitfalls of a software solution, but in a format and a cost point that would make it accessible to as many broadcasters as possible.
Having worked with the Raspberry Pi for SmartSign Lite, we were confident that this would provide the processing power needed to run a hardware encoder at a reasonable cost but that left some serious issues to overcome.
- Sound Input: The built in sound device in a stock Raspberry Pi is not really up to the task. The Pi is primarily intended to be used with an HDMI screen which will often include speakers. When using HDMI, the sound is sent digitally to the screen and it is the screen that decodes the audio. The built in audio chip on the Pi is not great quality and whilst it does have a 'microphone input' it is only accessible using a combined 4 pole mini jack which is a very unusual connector in radio. You could use an external USB sound device but this introduces a potential point of failure (the USB cable connecting them) and would not be a simple solution. Instead we sourced specialist sound device 'hat' for the Pi which offers high quality audio input and outputs as well as physical volume controls for both (and on common phono connectors). Solving the audio issue unfortunately led directly to the second issue.
- Enclosure: As you can see from the pictures above, the addition of the external sound device changes the size and shape of the Pi dramatically such that no 'off the shelf' Raspberry Pi enclosures would work. But we wanted to device that was simple to use and friendly to the end user. We also wanted to keep the price a s low as possible so fitting everything into a rack case and then extending all the connections out to external sockets was also not viable because of the martial and labour impact on the end cost. The answer was to custom design an enclosure that would incorporate all of necessary connections. Once the design was finalised, we decided that we would have them 3D printed to allow us to do smaller production runs and tweak the design if needed. To reduce 3D printing costs, enclosure was designed as a series of flat panels that fit together like a jigsaw to create the full case. Early models were printed in white and then sprayed purple but later versions had the colour added at time of print.
- Software: Once all of the physical challenges were addressed, we were able get back to our comfort zone and turn our attention to the software needed to make it all work. The StreamBox run a Linux core with all the necessary streaming services set to automatically start which means once it is configured, you can simply plug it in and it will start encoding and sending the stream to your service provider. We wanted the system to be as easy to setup as possible so our team developed a simple user interface for configuring the StreamBox and checking status as required. If you plug in a monitor, keyboard and mouse, you will see the StreamBox interface which allows you to alter the steam quality and format as well as add in your providers connection details.
So there you have it.
A simple 'Friday afternoon' project turned into a viable, cost effective solution. Full credit goes to John-Michael Sugden who has handled just about every element of this project.
The StreamBoox is available now, check our the StreamBox webpage
for more details.
Cloud Playout Phase 1 – Winding Down We initially launched Myriad Cloud Playout, now referred to as Phase 1, in the Autumn of 2020 to provide a solution for stations launching in the throes of a worldwide pandemic that rendered a more traditional approach to launching a radio station impractical. At the time we outlined a multi-phased approach to developing the technology in a way that would provide an immediate solution whist paving the way for wider adoption and deployment at scale. Phase 1 (Cloud Playout)– Desktop playout running on hardware hosted at Broadcast Radio HQ. Initially this involved Myriad Playout running on physical or virtual hardware with full remote access via RealVNC and voice tracking & live control via Myriad Anywhere. All other function (Myriad Schedule etc) was done via remote access. Phase 2 (Myriad Cloud Dedicated) – Desktop playout running on datacentre based VM’s offering higher levels of power and connectivity resilience. This was coupled with an extension of Myriad Anywhere to provide true web-based functionality for all major elements of Myriad including Schedule, Admin, User Management etc. Direct remote access is still included via RealVNC to allow for administration of 3rd party applications, Myriad Splits and BR Encoder. Phase 3 (Myriad Cloud Native) – Complete cloud solution running natively on MS Azure without the need for Windows 10/11 host VM. Control over all functionalities including encoding, audio processing and all Myriad functions provided via web interface. Includes native integration with streaming and web players etc. As there is no host Windows PC/VM, there is no direct remote access and no provision to run 3rd party applications. As part of out ongoing commitment to provide the best technical and cost-effective solutions for our customers, we will be deprecating Phase 1 services over the coming months with the option for customers to choose whether to switch to Myriad Cloud Dedicated or Myriad Cloud Native. Which should you pick? We have already identified which Phase 1 customers will likely need to switch Myriad Cloud Dedicated and which have the option to switch to Myriad Cloud Native instead, although the ultimate choice will be with you the customer. Broadly speaking the main factors to consider are: You should consider moving to Myriad Cloud Dedicated if: You are using Myriad Split Playout If you are using specialist 3rd party applications (audio processor, encoder, logging etc) If you are using Myriad Logging on the same PC/VM You want to retain direct remote access for other reasons You should consider moving to Myriad Cloud Native if: You want to avoid issues caused by Windows Updates or hardware driver issues. You want to be able to manage all aspects of your station (including encoding, streaming, archive etc) from a single web portal. Want to be able to add and remove stations and share content between stations managed via your portal. Get seamless integration with web player and optional website CMS / mobile apps. Want to benefit from the reliability and scalability of MS Azure. There are benefits to both platforms and we fully understand that whilst we can make a recommendation, it is down to the customers choice which is the best route for them. In either case, we will handle the migration for you so that all your content is replicated to the new platform and the transition is as seamless as possible. Phase 1 customers will receive emails in the coming days outlining the options and inviting them to contact us for more information and to confirm their decision. Phase 1 accounts for a tiny percentage of our overall Myriad Cloud (dedicated and native) customers so the vast majority of Myriad Cloud users will not be affected. If you have not received an email in the next ten days then you are already on Phase 2 or Phase 3 and the ramp down of Phase 3 will not impact you at all. As ever, if you have any questions, please contact the Broadcast Radio team for more information.
The body content of your post goes here. To edit this text, click on it and delete this default text and start typing your own or paste your own from a different source.
