Integrate ambivalence-structure analysis into your toolchain.

# Submit a passage for ambivalence-structure analysis
curl -X POST https://api.llulllab.com/asge/v1/analyse \
  -H "X-Session-Token: $ASGE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "text":     "The text to analyse. Paste a transcript, an interview, a speech, or any natural-language passage.",
    "language": "en",
    "mode":     "text"
  }'

Request fields

Field	Type	Description
text	string · required	The passage to analyse. Up to 10,000 words per submission. Line breaks segment temporal units.
language	enum · required	One of en, de, fr, es, tr, ja, he, fa.
mode	enum · optional	text (default) or microphone. The latter is reserved for sessions originating from a live recording context.
context	object · optional	Optional metadata kept alongside the case record (source, author, date, location). Up to 4 KB.

Response shape

{
  "success": true,
  "analysis": {
    "pairs": [
      { "minute": 3, "type": "classic",    "context": "..." },
      { "minute": 7, "type": "contextual", "context": "..." }
    ],
    "trend": [
      { "minute": 1, "level": "low", "note": "..." }
      // minute-by-minute
    ],
    "peak_minute":      4,
    "dominant_theme":   "freedom ↔ constraint",
    "conflict_summary": "Two to three sentences naming the sustained tensions.",
    "alert_level":      3
  },
  "stats": {
    "word_count":    1820,
    "pair_count":    24,
    "classic_count": 17,
    "context_count": 7,
    "pairs_per_min": 2.4
  }
}

Response fields

Field	Type	Description
success	boolean	True when the three-stage computational pipeline completed.
analysis.pairs	array	Each entry: minute, type (classic or contextual), and a short context fragment.
analysis.trend	array	Minute-by-minute density with level (low, medium, high) and a one-line analytical note.
analysis.peak_minute	integer	The minute index where ambivalence density is highest.
analysis.dominant_theme	string	The dominant semantic-conflict theme, expressed as an opposition.
analysis.conflict_summary	string	Two to three sentences naming the sustained tensions across the passage.
analysis.alert_level	integer	Graduated alert from 1 to 4. Higher values correspond to denser concentrations.
stats	object	Counts and densities: words, pairs, classic and contextual splits, pairs per minute.