Feature Request: Option to bypass audio processing and maintain original sample rate/bit depth

[epochaudio]

Title: Feature Request: Option to bypass audio processing and maintain original sample rate/bit depth

Description

I would like to request a feature to bypass all audio processing in Music Assistant and output the audio stream with its original sample rate and bit depth. This would allow for bit-perfect playback without any resampling or audio manipulation.

Use Case

• For users with high-end audio equipment who want to preserve the original audio quality
• For streaming high-resolution audio files without any downsampling
• For maintaining bit-perfect playback chain from source to output device

Proposed Solution

Add a new option in the stream controller settings:

• Add a "Bypass Processing" toggle in core settings
• When enabled:
  • Disable resampling
  • Disable volume normalization
  • Disable any audio effects/processing
  • Pass through the original sample rate
  • Maintain original bit depth
  • Keep original audio format (e.g., FLAC, DSD)

Technical Details

The implementation would require:

1. Identifying where the audio processing pipeline is in the codebase
2. Adding a bypass mode that skips the processing chain
3. Ensuring the original audio parameters are preserved throughout the streaming process
4. Adding appropriate UI controls in the settings

Benefits

• Better audio quality for audiophile setups
• Lower CPU usage (no processing overhead)
• True bit-perfect playback capability
• Support for high-resolution audio formats at native quality

Additional Context

This feature would be particularly useful for users who:

• Have DACs that handle their own processing
• Want to ensure no quality loss in their audio chain
• Need bit-perfect playback for specific use cases

Would this be something that could be considered for implementation?

[marcelveldt]

This is not the intended audience/usecase for Music Assistant.

Strictly taken we do close to no processing as soon as you disable volume normalization. The only thing we do is decode into PCM internally and that has to do with the fact we need to be compatible with a whole bunch of audio sources and target players - there is no loss of audio data at all.

Bits are bits. Audio quality is lost when you encode into a lossy format (or output to some shitty speakers), not in other ways.
Audiophiles that believe in these kind of myths can resort to software like Roon. We are more targeted to a broader audience.
Bit perfect is just marketing bullshit. Sorry to say it that way. I'm a believer of common sense and thats why/how I've built this all (and got me very far).