# Speaker Cue Crosswalk Snapshot Compare

> Speaker Cue Crosswalk Snapshot Compare is a paid API for AI agents from medien.halowerk.com, paid per call via x402, $0.031/call, status unknown (last checked 2026-10-25).

Cross-references subtitle speaker cues against source words and flags label mismatches by comparing a caller-supplied current state to a caller-supplied previous snapshot, returning exact change witnesses.

## Facts

- Endpoint: POST https://medien.halowerk.com/v1/baumarkt/media/speaker-cue-crosswalk/snapshot-compare?utm_source=zero.xyz
- Price: $0.031/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-25
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/speaker-cue-crosswalk-snapshot-compare-04a84c32
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_QYbNl7SgeARVoptYuYjHe

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 speaker-cue-crosswalk-snapshot-compare-04a84c32 -d '<json body>'
```

Example prompt: Compare my current subtitle speaker cue file against this previous snapshot and flag any mismatches where the speaker labels don't match the source words — I need to know exactly which cues changed.

## When to prefer this

Use this endpoint when you need to validate speaker cue label consistency between two versions of a subtitle file and get a precise diff of mismatches, without submitting your historical baseline to any third-party provider. Ideal for QA pipelines where subtitle accuracy between edit passes must be verified before payment or publication.

## Known failure modes

- Malformed or missing current cue data returns validation error
- Baseline snapshot format incompatible with current cue format
- No mismatches found returns empty change witness list
- Provider call fails for current media, returning provider error
- Snapshot too large exceeds bounded input limits

## How this service works

Compare Cross-reference subtitle cues with contained source words and flag label mismatch. against a bounded caller-supplied previous snapshot and return exact change witnesses. Baseline is caller supplied and is never authenticated as historical server output. Pure bounded timing/source-quote validation runs before payment. Exactly one current motor call, zero baseline provider calls. Full current provider disclosure retained; baseline is not submitted to the provider.

## Output

A JSON response listing exact change witnesses: which subtitle cues have speaker label mismatches relative to the caller-supplied baseline snapshot, with details on the specific source words and labels that diverged. No baseline data is submitted to the underlying provider; only the current state is processed by the provider.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "source": {
   "type": "object",
   "oneOf": [
    {
     "required": [
      "url"
     ]
    },
    {
     "required": [
      "content_base64"
     ]
    }
   ],
   "properties": {
    "url": {
     "type": "string",
     "pattern": "^https?://",
     "maxLength": 2048
    },
    "filename": {
     "type": "string",
     "maxLength": 260
    },
    "content_base64": {
     "type": "string",
     "maxLength": 12000000,
     "minLength": 4
    }
   },
   "additionalProperties": false
  },
  "baseline": {
   "type": "object",
   "required": [
    "words",
    "media"
   ],
   "properties": {
    "cues": {
     "type": "array",
     "items": {
      "type": "object"
     },
     "maxItems": 20000,
     "minItems": 1
    },
    "media": {
     "type": "object",
     "required": [
      "duration_seconds"
     ],
     "properties": {
      "duration_seconds": {
       "type": "number",
       "maximum": 1800,
       "exclusiveMinimum": 0
      }
     }
    },
    "words": {
     "type": "array",
     "items": {
      "type": "object",
      "required": [
       "word",
       "start",
       "end",
       "speaker"
      ],
      "properties": {
       "end": {
        "type": "number",
        "maximum": 1800.5,
        "exclusiveMinimum": 0
       },
       "word": {
        "type": "string",
        "maxLength": 3000,
        "minLength": 1
       },
       "start": {
        "type": "number",
        "maximum": 1800.5,
        "minimum": 0
       },
       "speaker": {
        "type": [
         "integer",
         "null"
        ],
        "minimum": 0
       },
       "confidence": {
        "type": [
         "number",
         "null"
        ],
        "maximum": 1,
        "minimum": 0
       }
      },
      "additionalProperties": false
     },
     "maxItems": 20000,
     "minItems": 1
    }
   }
  },
  "language": {
   "enum": [
    "de",
    "en",
    "es",
    "fr",
    "it",
    "nl",
    "pt",
    "multi"
   ],
   "type": "string"
  },
  "duration_minutes": {
   "type": "integer",
   "maximum": 30,
   "minimum": 1
  },
  "confidence_threshold": {
   "type": "number",
   "maximum": 1,
   "minimum": 0
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json"
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/speaker-cue-crosswalk-snapshot-compare-04a84c32/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from medien.halowerk.com](https://www.zero.xyz/host/medien.halowerk.com/llms.txt)