-
Notifications
You must be signed in to change notification settings - Fork 0
Reference Audio
VxSound includes four project-owned synthetic reference clips so EQ changes can be compared against repeatable material without bundling external recordings.
Designed to expose:
- low-end balance;
- midrange body;
- treble detail;
- stereo width;
- transient response;
- broad musical dynamics.
Designed to expose:
- left/right identification;
- moving pan;
- center-image stability;
- distance-style level/timbre cues;
- positional consistency.
It is a stereo localization reference, not a claim of true object-based or personalized HRTF rendering.
Contains original synthetic footstep-like transients, directional cues, low ambience and mechanical-style details. No game audio assets are used.
Contains original cinematic-style low-frequency energy, impacts, ambience, synthetic dialogue-band material and high-frequency detail.
Every generated clip is:
- 24 seconds;
- 48 kHz;
- 24-bit PCM;
- stereo WAV.
VxSound.Core/SyntheticTestAudio.cs generates the waveforms using oscillators, envelopes, deterministic procedural noise, filtering, panning and delays.
The clips are generated lazily and cached in %LOCALAPPDATA%\VxSound\test-audio. Large WAV binaries therefore do not need to live in Git history or the release archive.
Because generation is deterministic, the same material can be regenerated for repeatable comparisons and CI validation.
Selecting a reference button loads the generated WAV through the existing file-preview path:
reference WAV → active channel EQ → personal EQ → headroom → limiter → selected output
That makes the clips useful for comparing the actual profile and personal processing currently active in VxSound.
The automated test suite checks each generated reference for:
- expected WAV format;
- 48 kHz sample rate;
- 24-bit sample width;
- stereo channel count;
- expected duration/file size;
- non-silent signal.
Begin at a low comfortable level. Gaming and Movie material contains short transients and low-frequency energy. These references are not hearing-threshold tests.
See docs/TEST_AUDIO.md for the technical provenance statement.