Skip to main content

Introduction to HiBy Sankofa AI

HiBy Sankofa AI is a deep learning-based timbre cloning and simulation system. Its core is to use AI to learn the acoustic characteristics of classic audio equipment, then apply these "timbre characteristics" to the output of modern players in real time, allowing you to experience the sound of vintage devices such as MD players and Walkmans on modern equipment and restore a retro listening experience.

Core Positioning and Principles​

I. Core Positioning​

  • Full name: Sankofa AI Timbre Cloning System (debuted on HiBy RS8 II)
  • Goal: Recreate the unique timbre and listening experience of classic devices including MD, cassette, vinyl, and CD players, while retaining the power output, resolution, and dynamics of modern players
  • Name meaning: Sankofa originates from African culture, meaning "look back to the past and draw wisdom", corresponding to "reproducing classics and inheriting timbres"

II. Technical Principles​

1. Device Sampling and Data Collection​

  • Perform high-precision acoustic sampling on target classic devices (such as Sony MZ-R30, Sharp MS100 and other MD players)
  • Collected data: frequency response curve, harmonic structure, dynamic range, transient response, noise characteristics, distortion type and other complete acoustic fingerprint data
  • Method: professional test signals + real music materials, recording the output characteristics of devices from multiple dimensions

2. AI Model Training (Feature Extraction and Modeling)​

  • Use deep learning networks (CNN / Transformer and other audio-specific architectures) to learn the timbre fingerprint of devices
  • Core tasks:
    • Extract unique acoustic feature patterns of devices (such as softened high frequencies of MD, saturation distortion of cassettes, low-frequency texture of vinyl records)
    • Establish a mapping model from "input signal → target device output"
    • Filter out inherent defects of vintage devices (such as insufficient power, compressed dynamics of old MD players) and retain only timbre charm

3. Timbre Library and Model Deployment​

  • Trained models are packaged as timbre modules or timbre libraries (such as MD timbre pack, cassette timbre pack)
  • Stored in the player (RS8 II) and can be loaded or switched at any time

4. Real-Time Timbre Rendering (Playback Application)​

  • During music playback, Sankofa AI processes the digital audio stream in real time
  • Key feature: real-time, low-latency processing without affecting playback smoothness

Simulated Audio Equipment​

Note: The following lists all devices currently supported by Sankofa AI, with clear labels for gain range, Deep BASS support, etc., for quick reference and to avoid confusion about missing functions. Continuously updated…

Sankofa AI Supported TimbreGain Adjustment RangeDeep BASS Support
Sharp MS100-12~0 dBSupported, with BASS1, BASS2, BASS3
Sharp MT831-12~0 dBSupported, with BASS1, BASS2, BASS3
Sharp DR7-12~0 dBSupported, with BASS1, BASS2, BASS3
Sony MZ-R30-12~0 dBSupported, with BASS1, BASS2
Sony MZ-R50-12~0 dBSupported, with BASS1, BASS2
Sony MZ-NH1-12~0 dBNot supported
Sony MZ-E730-12~0 dBNot supported
Panasonic SJ-MJ100-12~0 dBNot supported

Where to Find & How to Use​

HiBy Player (Android)​

Settings → Sankofa AI

Supported Devices​

  • HiBy RS8II

Frequently Asked Questions​

  1. Why can't I find a certain function for a simulated device? Supported functions vary for each simulated device and are clearly marked in the table above. Unmarked functions are not supported by that device, which is not a malfunction.

  2. How to use Sankofa AI? Open the player, find the "Sankofa AI" entry, select the device you want to simulate to switch timbres. Supported adjustments (bass, volume, etc.) can be set in the corresponding device options.

  3. Do all simulated devices support the same functions? No. Different simulated devices have different core timbres and supported adjustment functions. Please refer to the table above; there is no need to search repeatedly for unsupported functions.