pith. sign in
Pith Number

pith:7ECUX2WT

pith:2026:7ECUX2WT3FVS2IHK4OZBKP6AL6
not attested not anchored not stored refs pending

Explainable AI in Speaker Recognition -- Making Latent Representations Understandable

Mark D. Plumbley, Wenwu Wang, Yanze Xu

Speaker recognition neural networks organize their latent representations into hierarchical clusters that align with semantic attributes like gender and nationality.

arxiv:2604.23354 v2 · 2026-04-25 · eess.AS · cs.AI · eess.SP

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{7ECUX2WT3FVS2IHK4OZBKP6AL6}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp
2 Internet Archive
3 Author claim open · sign in to claim
4 Citations open
5 Replications open
Portable graph bundle live · download bundle · merged state
The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

This work applies SLINK and HDBSCAN to demonstrate the existence of hierarchical clustering phenomena within the network representation space, and designs HCCM to perform one-to-one matching between predefined semantic classes and hierarchical representation clusters, with Liebig's score to quantify performance.

C2weakest assumption

That the hierarchical clusters produced by SLINK or HDBSCAN correspond to meaningful semantic classes or their conjunctions in a non-arbitrary way that HCCM can reliably detect and that Liebig's score meaningfully diagnoses limiting factors.

C3one line summary

Speaker recognition networks form hierarchical clusters in latent space that can be matched to semantic classes using new HCCM algorithm and quantified by Liebig's score.

Receipt and verification
First computed 2026-05-29T01:05:10.514741Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

f9054bead3d96b2d20eae3b2153fc05f965498ef02a3c8ce2306eff1901e3ba4

Aliases

arxiv: 2604.23354 · arxiv_version: 2604.23354v2 · doi: 10.48550/arxiv.2604.23354 · pith_short_12: 7ECUX2WT3FVS · pith_short_16: 7ECUX2WT3FVS2IHK · pith_short_8: 7ECUX2WT
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/7ECUX2WT3FVS2IHK4OZBKP6AL6 \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: f9054bead3d96b2d20eae3b2153fc05f965498ef02a3c8ce2306eff1901e3ba4
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "b846fd5324a4735d8795f09020c794678597c233ede5167750785e06b97572d1",
    "cross_cats_sorted": [
      "cs.AI",
      "eess.SP"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "eess.AS",
    "submitted_at": "2026-04-25T15:44:20Z",
    "title_canon_sha256": "ae83f33cba4c15379fe40b1661e10e8ff8fdb173de9ca753b34d66df75b37c45"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.23354",
    "kind": "arxiv",
    "version": 2
  }
}