Post AstJ2wSJEPUnsCkfVA by manawyrm@chaos.social
 (DIR) More posts by manawyrm@chaos.social
 (DIR) Post #AstJ2wSJEPUnsCkfVA by manawyrm@chaos.social
       2025-04-08T14:55:40Z
       
       0 likes, 0 repeats
       
       New blog post:Speaker DSP for a tiny 9" notebook running Linux (using PipeWire, similar to Asahi), GPD Pocket 4:https://kittenlabs.de/blog/2025/04/06/gpd-pocket-4-speaker-dsp/https://www.youtube.com/watch?v=wNm5xq_xw3s
       
 (DIR) Post #AstJ2xiIYRsbm5Ex3w by wolf480pl@mstdn.io
       2025-04-08T19:38:43Z
       
       0 likes, 0 repeats
       
       @manawyrm noob question, did I understand the maths correctly? :1. you measured frequency response of the speaker (a function of type freq -> amplitude)2. you did y(f) = 1 / x(f), which in log scale looks like inverting the dB axis3. you did inverse FFT to convert it from y(f) to y(t), which is called impulse response, saved as a wav file4. you told the DSP to replace every sample (impulse) with that wav file (impulse response) multiplied by the sampleis that right?
       
 (DIR) Post #AstJ7eLfypgvABtUhc by wolf480pl@mstdn.io
       2025-04-08T19:39:37Z
       
       0 likes, 0 repeats
       
       @manawyrm also, is 4. equivalent to using the impulse response samples as coefficients of a FIR filter?
       
 (DIR) Post #AstJYeHeKV8EmSsups by manawyrm@chaos.social
       2025-04-08T19:44:26Z
       
       0 likes, 0 repeats
       
       @wolf480pl I'm not a DSP expert at all, but yes, as far as I understand it, that's  it.
       
 (DIR) Post #AstJiyftN4nguz9JTc by wolf480pl@mstdn.io
       2025-04-08T19:46:21Z
       
       0 likes, 0 repeats
       
       @manawyrm thanks, now I understand where FIR filter coefficients come from