Download Zero-Phase Sound via Giant FFT Given the speedy computation of the FFT in current computer
hardware, there are new possibilities for examining transformations for very long sounds. A zero-phase version of any audio
signal can be obtained by zeroing the phase angle of its complex
spectrum and taking the inverse FFT. This paper recommends additional processing steps, including zero-padding, transient suppression at the signal’s start and end, and gain compensation, to
enhance the resulting sound quality. As a result, a sound with the
same spectral characteristics as the original one, but with different temporal events, is obtained. Repeating rhythm patterns are
retained, however. Zero-phase sounds are palindromic in the sense
that they are symmetric in time. A comparison of the zero-phase
conversion to the autocorrelation function helps to understand its
properties, such as why the rhythm of the original sound is emphasized. It is also argued that the zero-phase signal has the same
autocorrelation function as the original sound. One exciting variation of the method is to apply the method separately to the real
and imaginary parts of the spectrum to produce a stereo effect. A
frame-based technique enables the use of the zero-phase conversion in real-time audio processing. The zero-phase conversion is
another member of the giant FFT toolset, allowing the modification of sampled sounds, such as drum loops or entire songs.
Download Audio-to-Audio via Diffusion Warm Initialization In this paper, we propose diffusion warm initialization as a simple yet effective approach for a range of audio-to-audio transformation tasks. To illustrate the generality of the approach, we demonstrate its use in timbre transfer, MIDI-to-Real synthesis, and multiple audio enhancement tasks. We conduct a detailed empirical analysis on timbre transfer to investigate the role of the initialization time t_init. The effect of t_init is evaluated using pitch-based Jaccard Distance and Fréchet Audio Distance to quantify faithfulness to the input signal and alignment with the target distribution. Our results provide practical guidance for selecting t_init and show that, once properly chosen, a single pretrained diffusion model combined with warm initialization can support multiple transformation objectives without task-specific training or conditioning. Despite its simplicity, this approach already achieves competitive results when compared with more complex pipelines designed specifically for these tasks.
Download Compiling Differentiable Audio Graphs to Real-Time DSP Differentiable audio processors are habitually designed and optimised in machine-learning frameworks, but deploying them as real-time audio effects still often requires non-automatic implementation in a dedicated digital signal processing language. The translation is error-prone, demands an onerous verification process, and detaches research prototypes from usable production tools. That being so, we present ADAC, a compiler that lowers a trained model to a framework-agnostic intermediate representation and emits efficient FAUST code whose impulse response matches the source model to within floating-point arithmetic noise, direct paths included. The optimisation loop is made audible by replacing the model in a running plugin after each gradient step. The exported processor carries a small set of macro-controls that leave its stability intact. A stability certificate computed from the shipped parameters is checked before the plugin is built. At the demonstration, a feedback delay network is trained and exported to a working plugin.