# Halowerk Voice Clone Check — Embedding Similarity Scorer

> Halowerk Voice Clone Check — Embedding Similarity Scorer is a paid API for AI agents from creator.halowerk.com, paid per call via x402, $0.003/call, status unknown (last checked 2026-09-13).

Computes a weighted similarity score between two numeric voice embeddings combined with caller-supplied spectral distance, pitch-contour correlation, and timing distance to produce a composite voice-match score.

## Facts

- Endpoint: POST https://creator.halowerk.com/v1/voice-clone-check
- Price: $0.003/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-13
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/halowerk-voice-clone-check-embedding-similarity-scorer-2b1cfd2a
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_CSJGRv19GpyUxnD1xtUmv

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability halowerk-voice-clone-check-embedding-similarity-scorer-2b1cfd2a -d '<json body>'
```

Example prompt: Score how similar these two speaker embeddings are: my reference embedding is [0.12, -0.45, 0.78, ...] and the candidate is [0.11, -0.43, 0.80, ...], with spectral distance 0.2, pitch-contour correlation 0.85, and timing distance 0.15.

## When to prefer this

Use this endpoint when you have already extracted numeric voice embeddings and supplementary feature distances (spectral, pitch, timing) externally and need a fast, deterministic weighted composite similarity score. Do not use it to process raw audio, identify speakers, detect clones, assess consent, or flag fraud — it performs only numeric aggregation of caller-supplied features.

## Known failure modes

- Mismatched embedding lengths between candidate and reference vectors return a validation error
- Values outside allowed ranges (e.g. spectral_distance_0_1 > 1 or pitch_contour_correlation < -1) are rejected
- Embeddings with fewer than 2 or more than 4096 elements are rejected
- Missing required fields result in an error response
- Numeric overflow from extreme embedding values near ±1,000,000 may cause computation errors

## How this service works

Calculates cosine similarity between caller-supplied numeric embeddings and combines it with supplied spectral distance, pitch-contour correlation and timing distance using fixed weights. It processes no audio and cannot establish speaker identity, consent, cloning, authenticity or fraud.

## Output

Returns a composite voice-match score derived from the cosine similarity between the two supplied numeric embeddings, combined with the caller-provided spectral distance, pitch-contour correlation, and timing distance using fixed internal weights. Does not process audio, infer speaker identity, or determine authenticity.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "candidate_embedding": {
   "type": "array",
   "items": {
    "type": "number",
    "maximum": 1000000,
    "minimum": -1000000
   },
   "maxItems": 4096,
   "minItems": 2
  },
  "reference_embedding": {
   "type": "array",
   "items": {
    "type": "number",
    "maximum": 1000000,
    "minimum": -1000000
   },
   "maxItems": 4096,
   "minItems": 2
  },
  "timing_distance_0_1": {
   "type": "number",
   "maximum": 1,
   "minimum": 0
  },
  "spectral_distance_0_1": {
   "type": "number",
   "maximum": 1,
   "minimum": 0
  },
  "pitch_contour_correlation": {
   "type": "number",
   "maximum": 1,
   "minimum": -1
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/halowerk-voice-clone-check-embedding-similarity-scorer-2b1cfd2a/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from creator.halowerk.com](https://www.zero.xyz/host/creator.halowerk.com/llms.txt)