Building a DSP template

A DSP template isn’t something I pull off a shelf. Every one starts with the client and the space: who’s using the room, what they need to hear, and what the room will do to the sound.

Start with the I/O. Before I open the software, I list every input and output: microphones, program sources, codecs, amplifier channels, recording and broadcast feeds. That list is the backbone of everything that follows, and it’s the first thing I check when something doesn’t add up later.

Map the routing before assigning anything. Once I know what comes in and what goes out, I work out where each signal needs to go and when. In a small room that’s quick. In a larger system with divisible spaces, presets for different events, or routing that changes with how the room is used, I document the routing before I assign a single block. A conflict is much cheaper to find in a document than in a running system.

Build the processing chain for the space. Then come the tools: acoustic echo cancellation where there’s conferencing, filtering and EQ, delays to time-align speaker zones, automatic mixing for multiple open microphones, and dynamics where they help. None of it is there by default. Each block earns its place because the room or the use calls for it.

Let the room have the final word. A template gets you to site; the room finishes it. Measurements taken during commissioning feed back into the filters, delays and levels, so the final settings reflect the real space rather than the drawing.

Record everything. Every routing decision, preset and change made on site gets written down. Whoever opens the file next, including me two years later, should be able to see what the system does and why without reverse-engineering it.