AVDP Synthetic Corpus

Synthetic Hebrew audio clips for the Audio Violence Detection Pipeline. Hebrew (he-IL) · 16 kHz mono 16-bit PCM · generated by SynthBanshee.

delivery 003 · 2026-05-12 · provisional 20 clips · ~41.6 min · she_proves (12) + elephant_in_the_room (8) · Azure (18) + Google (2) · 0 validation failures.

---

See it first

A real clip from the corpus — Severe Violence scene, two speakers, with strong-label events overlaid on the waveform:

You can read the typical escalation arc directly: an argument starts as verbal (VERB, blue), peaks into distress vocalisations (DIST, orange) around 36s, then into physical-violence cues (PHYS, red) around 71s. Intensity badges (I2 → I5) follow the same curve.

---

Load a clip in 4 lines

import json, soundfile as sf
wav, sr = sf.read("data/he/agg_m_30-45_001/sp_sv_a_0001_00.wav")
meta    = json.loads(open("data/he/agg_m_30-45_001/sp_sv_a_0001_00.json").read())
print(f"{len(wav)/sr:.1f}s  has_violence={meta['weak_label']['has_violence']}  "
      f"intensity_max={meta['weak_label']['max_intensity']}")
# 110.5s  has_violence=True  intensity_max=5

For everything else: Start here →

---

Two consumer teams

smartphone app

She-Proves

Passively monitors a phone for domestic-violence incidents and preserves audio evidence for legal use. High-recall orientation — better to flag and review than to miss.

12 clips · Tier A (clean audio) · scenes 3–6 min · phone-pocket device profile.

She-Proves guide →

raspberry pi · clinic / welfare office

Elephant in the Room

A Pi-class device that alerts security when a social worker is under threat. High-precision orientation — false alarms erode trust.

8 clips · Tier B (room IR + budget mic + noise) · scenes 1–4 min · alert in final 40%.

Elephant guide →

---

Where to go

First time here	Start here — clone, load one clip, read its labels
About to write code	Common mistakes — read this once; it'll save you a few
Decoding a label	Label Taxonomy — typologies, categories, has_violence rule
Decoding a JSON field	Schema Reference — annotated ClipMetadata example
Working with team data	She-Proves · Elephant
Looking up a term	Glossary — F0, SSML, IR, AGG/VIC/SW/BEN, etc.
Checking what's current	Deliveries — current batch, known gaps

---

This is a small test batch, not training data

All current deliveries are preview batches for verifying that downstream data-loading code works before the full dataset arrives. The split column in manifest.csv is informational only — all 20 clips are split: train because there aren't enough unique speakers for a disjoint partition at this scale. Do not train production models on this corpus.

What's not in this corpus

No real human recordings (synthetic TTS only) · no Arabic or English (Hebrew only) · no inter-annotator agreement metrics (labels are auto-generated by SynthBanshee) · no demographic detail beyond gender + age_range. Scripts are LLM-generated in Hebrew, not human-written. See Glossary for what each abbreviation means.