ASR Open Source Streamer Project

[StefaanE]

Thanks @amirm! Great to be on the team. I’m more of a software development and project management guy, so I’ll be happy with suggestions, links and other material relevant to the requirements Amir outlined. Once we have clarity in that department, we can make a decision on the road to take. I’m still in full-time employment, but I promise to do a daily summary, and I’ll try to be as reactive as possible.

 

[voodooless]

I would seriously suggest finding an answer to the "why" and "what" questions before going into this endeavour. Why is this project needed? Why will this be different from the others out there? What is the reason for yet another open-source streamer? Why not just hop on to whatever is there to expand and improve it?

 

[StefaanE]

I would seriously suggest finding an answer to the "why" and "what" questions before going into this endeavour. Why is this project needed? Why will this be different from the others out there? What is the reason for yet another open-source streamer? Why not just hop on to whatever is there to expand and improve it?

It's not a commercial project, so the 'why' can be simply -- because we love to do it. But I think that if we find something that ticks almost all the boxes, and can be improved, it'll certainly be a favoured solution. There are already too many open doors being kicked in anyway (but kicking in open doors is harmless and can be fun too ).

 

[Piranesi]

This is a wonderful idea. I've been using Moode (moved to it from Volumio) and while wonderful it is also buggy as shit.

All that I think myself and many others need is a simple, stable streaming endpoint with DSP (i.e. pEQ and room correction).
Web radio, screen support, scrobbling etc- completely unecessary.

edit: the suggested stream modules look very suitable, but availability might be difficult outside the EU. The advantage of RPi is easy availability pretty much anywhere.

 

[voodooless]

It's not a commercial project, so the 'why' can be simply -- because we love to do it.

I'd call it: YAPS - Yet Another Pi Streamer

Don't get me wrong here, I like the incentive very much. But there must be a good reason to start this, other than "loving to do it". If it doesn't add anything to the other options, it's likely not going to be very successful.

But I think that if we find something that ticks almost all the boxes, and can be improved, it'll certainly be a favoured solution.

I would suggest that that is the best way to approach this if you want fast results. You could also consider a fork of an existing project.

Regarding a case, I think there is quite some room for improvement there. Especially cases that are kind of HiFi looking, with a nice integrated touchscreen, proper rotary encoder and nice buttons, IR remote, options for various HAT's and extensions... All for a fair price . But you might want to split it into two separate projects then probably.

 

[Mnyb]

Squeezelite endpoint (Squeezebox emulation/software player ) ? so that folks like me with a LMS server can use it

 

[somebodyelse]

I second taking a good look at why this shouldn't be a plugin or improvements to the best fit existing player project. They already tick many (most) of the boxes. Is there something that rules out improving an existing project rather than starting from scratch?

Proprietary compatibility issues
As has been pointed out already, there are no open implementations of Airplay2 or the ChromeCast endpoint. Roon currently distribute a binary version of their endpoint that will run on the Pi and probably many other ARM SBCs, but they could withdraw that whenever they want. I understand Roon can also stream to squeezebox endpoints, so squeezelite has that covered.

On the crossover front:
The Pi and most of the widely available SBC options only support 2 channels of audio via I2S, which is why you won't find generic multichannel DAC hats. Someone managed to abuse it into doing 8 channels for a research project, but it was unreliable, with channels 'slipping' from time to time - so your sub channel might suddenly end up going to your tweeter. The few that do have >2 output channels aren't independent - they're receiving 2 channels and sending them to >2 outputs, usually with some DSP involved. The Allo Piano 2.1 uses TI's onboard DSP in the DACs, while HifiBerry use AD's SIgmaDSP.

The multichannel USB interface option has its own issues - while hifi style stereo DAC linux compatibility is almost certain, multichannel is less so. We would need to identify some interfaces with good performance and compatibility.

We could pick one of the few SBCs with TDM capability (Beaglebone Black, perhaps others?) but then run into the lack of available and tested output methods. Also the Beaglebone has a relatively weak CPU which may rule out its use for DSP.

Every time I look at this the cost effective option seems to be to buy FusionAmps - if only there was a linux-based interface to the DSP.

 

[StefaanE]

I'd call it: YAPS - Yet Another Pi Streamer

-- I was thinking of "Noah" -- it's not a stream, it's a flood.

 

[StefaanE]

Regarding a case, I think there is quite some room for improvement there. Especially cases that are kind of HiFi looking, with a nice integrated touchscreen, proper rotary encoder and nice buttons, IR remote, options for various HAT's and extensions... All for a fair price . But you might want to split it into two separate projects then probably.

Is it a software project, or a mixed software-hardware project? Including specific hardware makes for a nicer result, but also might limit availability outside of certain "favoured" markets, and I don't know if ASR can/should become involved in things like group purchases etc. These things tend to become very complex (been there, got the t-shirt).

 

[StefaanE]

I second taking a good look at why this shouldn't be a plugin or improvements to the best fit existing player project. They already tick many (most) of the boxes. Is there something that rules out improving an existing project rather than starting from scratch?

Proprietary compatibility issues
As has been pointed out already, there are no open implementations of Airplay2 or the ChromeCast endpoint. Roon currently distribute a binary version of their endpoint that will run on the Pi and probably many other ARM SBCs, but they could withdraw that whenever they want. I understand Roon can also stream to squeezebox endpoints, so squeezelite has that covered.

On the crossover front:
The Pi and most of the widely available SBC options only support 2 channels of audio via I2S, which is why you won't find generic multichannel DAC hats. Someone managed to abuse it into doing 8 channels for a research project, but it was unreliable, with channels 'slipping' from time to time - so your sub channel might suddenly end up going to your tweeter. The few that do have >2 output channels aren't independent - they're receiving 2 channels and sending them to >2 outputs, usually with some DSP involved. The Allo Piano 2.1 uses TI's onboard DSP in the DACs, while HifiBerry use AD's SIgmaDSP.

The multichannel USB interface option has its own issues - while hifi style stereo DAC linux compatibility is almost certain, multichannel is less so. We would need to identify some interfaces with good performance and compatibility.

We could pick one of the few SBCs with TDM capability (Beaglebone Black, perhaps others?) but then run into the lack of available and tested output methods. Also the Beaglebone has a relatively weak CPU which may rule out its use for DSP.

Every time I look at this the cost effective option seems to be to buy FusionAmps - if only there was a linux-based interface to the DSP.

Thanks for this. I have a lot of reading to do, it seems.

 

[NYfan2]

Great idea because this is what I'm looking for at the moment.

My preference is to integrate Spotify connect and Tidal connect. Because I prefer to use the native apps.
And I prefer USB out too my DAC

 

[StefaanE]

But there must be a good reason to start this, other than "loving to do it". If it doesn't add anything to the other options, it's likely not going to be very successful.

That's Open Source in a nutshell, unfortunately.

 

[mr.k]

While I think open source streamer is kewl idea, coming from software development business, I believe it's hell of the work to make everything work starting from scratch. I agree it would be the best to take one of the more popular distributions, and do some kind of UI/functional "make over" .

I recently moved from Volumio to Moode for two reasons: GUI simplicity and Graphic/Parametric equalizer.
Yes ,bugs... bugs everywhere! E.g. Spotify doesn't work in combination with graphic equalizer and apple dongle (USB out).

But, what I'm really missing from this Raspberry/DIY scene are nice boxes with nice (capacitive) screens
I'm not talking about generic AliExpress metallic boxes where you still need to do a lot of work buying components, screens, connectors, whatnot... Also not talking about cheap plastic/lego cases for official 7 inch raspberry screen.

E.G. few weeks ago, when new Gustard M16 streamer was mentioned (https://www.audiosciencereview.com/...242bc-be07-49f2-844e-6ef78cb9bc62-jpeg.67152/), my first thought was: it would be cool if I could buy only box and screen and put raspberry in

I'm really surprised how come nobody offers nice, living room/family friendly, Raspberry plug&play box with touch screen for control... someone like @March Audio, Nord, Apollon...

P.S. I own Flirc case, with Flirc IR control over Logitech harmony - and my wife still finds it complicated because "there are no buttons to press!", therefore my rant

 

[voodooless]

Is it a software project, or a mixed software-hardware project?

You're in charge it seems, so you answer the questions I just commented on one of @amirm points on the list about hardware While both are interesting things, I'd focus on one thing only. Only a software project is already quite a handful as you might already know.

On the crossover front:
The Pi and most of the widely available SBC options only support 2 channels of audio via I2S, which is why you won't find generic multichannel DAC hats. Someone managed to abuse it into doing 8 channels for a research project, but it was unreliable, with channels 'slipping' from time to time - so your sub channel might suddenly end up going to your tweeter. The few that do have >2 output channels aren't independent - they're receiving 2 channels and sending them to >2 outputs, usually with some DSP involved. The Allo Piano 2.1 uses TI's onboard DSP in the DACs, while HifiBerry use AD's SIgmaDSP.

I use the HifiBerry DAC + DSP to get more channels. It works pretty well, but obviously, you cannot do more than two channels of software filtering on the PI. The DSP is quite capable however, as long as you don't need any high tap count FIR it will do for most people. Also would love to see one or two extra SPDIF output (they can be hacked on though).

The multichannel USB interface option has its own issues - while hifi style stereo DAC linux compatibility is almost certain, multichannel is less so. We would need to identify some interfaces with good performance and compatibility.

Yeah, blame the manufacturers Some of the Pro interfaces have proprietary drivers and do not offer Linux support. It would definitely be good to have a list of such devices!

 

[StefaanE]

You're in charge it seems, so you answer the questions

At this stage I'm looking for input. I hope to find a couple of cats to herd that are far cleverer than I .

 

[jannek]

Overall, the RPi4 running volumio is almost exactly what is described in the first post.

- Hardware ist an RPi 4 with 4 Gb RAM and an Argon One Case for passive cooling.
- This is connected to a Topping D10s via USB. This DAC feeds only the headphone amplifier with analog signals, the digital signal is passed on to some active speakers with Hypex Fusion amps via S/PDIF.
- Speaker DSPs do software crossover, EQ , gain of the chassis and running time correction
- Inside Volumio two plugins are used:
- A software simulation of crossfeeds for headphone listening (some tracks really require this)
- some BruteFIR presets do headphone EQ or room EQ for speakers
- As sources local files on the NAS, internet radio and streaming of spotify are available. Tidal and Qobuz are available as extras, but not for free
- If the spotify integration is not liked (it's quite responsive and not so bad in my opinion) also Spotify connect works
- User interface is the volumio app on any Android tablet

What is missing at this setup?

 

[Glasvegas]

Overall, the RPi4 running volumio is exactly what is described in the first post.

- Hardware ist an RPi 4 with 4 Gb RAM and an Argon One Case for passive cooling.
- This is connected to a Topping D10s via USB. This DAC feeds only the headphone amplifier with analog signals, the digital signal is passed on to some active speakers with Hypex Fusion amps via S/PDIF.
- Speaker DSPs do software crossover, EQ , gain of the chassis and running time correction
- Inside Volumio two plugins are used:
- A software simulation of crossfeeds for headphone listening (some tracks really require this)
- some BruteFIR presets do headphone EQ or room EQ for speakers
- As sources local files on the NAS, internet radio and streaming of spotify are available. Tidal and Qobuz are available as extras, but not for free
- User interface is the volumio app on any Android tablet

What is missing at this setup?

Tidal integration is poor. On line streaming is the only music source for some members.

 

[jannek]

Heard that already. But with Spotifi HiFi about to come there is no reason to insist on Tidal anymore. Apart from the question if the difference between lossless FLAC and 320 kbps OGG is audible.

 

[voodooless]

At this stage I'm looking for input. I hope to find a couple of cats to herd that are far cleverer than I .

Don't worry, I wasn't trying to suggest that you'll need to answer all questions on your own

 

[bluefuzz]

I'm really surprised how come nobody offers nice, living room/family friendly, Raspberry plug&play box with touch screen for control.

I'm really surprised why anyone would want this. The advantage of a headless device is that it can be accessed on any device on your network: pc, telephone, iPad etc. A box with a touch screen would inevitably be tethered to the hifi by a physical cable so you would have to get up off your chair to access the touch screen. All the disadvantages of vinyl without even the euphonious distortion ... ;-)