SSL Bus Compressor Control Voltage Dataset

Name: SSL Bus Compressor Control Voltage Dataset
Creator: University of Rochester
Published: 2026-03-31 12:43:04
License: 暂无描述

DataCite Commons2026-03-31 更新2026-05-07 收录

下载链接：

https://rochester.figshare.com/articles/dataset/SSL_Bus_Compressor_Control_Voltage_Dataset/31892026

下载链接

链接失效反馈

官方服务：

资源简介：

<pre>--- # SSL Bus Compressor Control Voltage Dataset This repository hosts a dataset for the analysis and modeling of dynamic range compression in an SSL-style bus compressor. The dataset includes paired input/output audio signals along with the corresponding control signal applied to the voltage-controlled amplifier, expressed directly as gain reduction in decibels. This dataset is introduced in a forthcoming DAFx paper. --- ## Status This repository has been prepared in advance of the dataset release. The full dataset will be made available following publication. --- ## Overview The Solid State Logic bus compressor applies a time-varying gain g[k] to an input signal. In conventional datasets, only the input and output audio are available, requiring evaluation via proxy metrics defined on the waveform. This dataset includes the control signal corresponding to the applied gain, enabling direct computation of error with respect to the underlying quantity of interest. Unlike conventional datasets, this enables both system identification approaches and black-box or generative models to optimize directly for the applied gain trajectory rather than relying on proxy loss functions. The control signal is provided as gain reduction in decibels: G[k] proportional to CV[k] and has been pre-scaled such that it directly represents gain reduction in dB. --- ## Dataset Summary * **219** stereo input clips (30 seconds each) * **90** compressor parameter permutations * **19710** total input–output–control examples * Sample rate: **44.1 kHz** * Audio resolution: **24-bit** * Control signal resolution: **32-bit** Each input clip is processed under every control permutation. --- ## Contents The dataset includes: * Stereo input audio signals * Corresponding processed output signals from hardware * Control signals (CV) expressed as gain reduction (dB) * Calibration signals for static and dynamic analysis * Metadata describing compressor settings, derived parameters, and signal relationships --- ## Data Organization All relationships between signals are defined explicitly in metadata tables. ### Primary table * `metadata_main.parquet` Defines all clip–permutation pairs and provides paths to: * input audio * output audio * control signal (gain reduction) Each row corresponds to a single example. --- ### Supporting tables * `metadata_clips.parquet` Clip-level metadata (source, normalization, indexing) * `metadata_permutation.parquet` Compressor parameter configurations (attack, release, ratio, etc.) * `metadata_comp_vals.parquet` Component values and derived time constants for circuit-informed modeling * `metadata_cal_inputs.parquet` Definitions of calibration input signals * `metadata_cal.parquet` Calibration outputs across permutations * `cc_calibration_summary.parquet` Summary of control constant derivation for CV[k]->G[k] conversion --- ### File structure * `audio_in/` — input clips * `audio_out/pXXX/` — processed outputs per permutation * `cv/pXXX/` — control signals (gain reduction) per permutation * `calibration_signals/audio_in` — calibration inputs * `calibration_signals/audio_out/pXXX/` — calibration audio outputs per permutation * `calibration_signals/cv/pXXX/` — calibration control signals (gain reduction) per permutation * `metadata/` — parquet tables The filesystem layout is not required to infer relationships between signals. All mappings are defined in metadata. --- ## Calibration Signals Calibration signals are included to support analysis of: * Static compression curve (ramp signal) * Attack and release behavior (burst signals) * System delay and alignment (MLS signal) * Control signal scaling The control signals in the dataset have already been scaled to represent gain reduction in dB. --- ## Important Notes * Control signals are stored as WAV files but are **not audio signals** * They may contain large DC components and are not bounded to [-1, 1] * These signals should not be routed to audio playback systems --- ## Example Usage ```python import pandas as pd import soundfile as sf df = pd.read_parquet("metadata/metadata_main.parquet") row = df.iloc[0] x, _ = sf.read(row.audio_in_path) y, _ = sf.read(row.audio_out_path) g, _ = sf.read(row.cv_path)``` --- ## Intended Use This dataset is intended for: * Behavioral modeling of dynamics processors * Control signal inference * Evaluation of loss functions for compressor modeling * Circuit-informed and gray-box modeling approaches * Black-box and generative modeling approaches trained directly on gain trajectories --- ## License This dataset is released under the CC-BY 4.0 license. --- ## Citation If you use this dataset, please cite the associated DAFx publication (to be updated upon release).</pre>

提供机构：

University of Rochester

创建时间：

2026-03-30

5,000+

优质数据集

54 个

任务类型

进入经典数据集