{
  "contents": [
    {
      "role": "user",
      "parts": [
        {
          "text": "Four excerpts of the SAME scene from four sung story-songs (one English, three Ukrainian). Listen to each against its own drum groove. For each attachment: grooveLock (how consistently the STRESSED syllables of the words land exactly on the drum beats or eighths, 1-5), earlyLate[{localSeconds,words,early|late}] where a phrase starts noticeably early or late or a stressed syllable falls between beats, smeared[{localSeconds,words}] where a long word is smeared across beats. Then rankTightestFirst: [attachment numbers] — which one sits tightest on the beat, down to the loosest — with one sentence each. Rank rhythm only, not language, voice or production. Return ONLY JSON (no prose outside JSON), <=1900 words, English. Required keys: mediaAccess(boolean), audioAccess(boolean), attachmentsHeard[{attachment,firstWordsHeard,lastWordsHeard,durationEstimateSeconds}], then the analysis keys listed. Quote only words you actually hear; mark uncertain words with (?). Use attachment number plus LOCAL seconds of that attachment. Do not assume any language or attachment is better a priori; do not reward louder mastering, more notes or a softer timbre. Identify concrete mechanisms (where the stressed syllables fall against the kick/snare, early or late phrase entries, words smeared across beats), not taste."
        },
        {
          "text": "Attachment 1: Attachment 1; parent A-master.wav; EXACT parent range [81.83, 96.83]; local audio starts 0. Use attachment number and local seconds, never concatenate timestamps. Fixed gain, no retiming."
        },
        {
          "inlineData": {
            "mimeType": "audio/mpeg",
            "data": "[base64 of frozen bytes; see inputs.json]"
          }
        },
        {
          "text": "Attachment 2: Attachment 2; parent v2-master.wav; EXACT parent range [76.56, 90.63]; local audio starts 0. Use attachment number and local seconds, never concatenate timestamps. Fixed gain, no retiming."
        },
        {
          "inlineData": {
            "mimeType": "audio/mpeg",
            "data": "[base64 of frozen bytes; see inputs.json]"
          }
        },
        {
          "text": "Attachment 3: Attachment 3; parent B-master.wav; EXACT parent range [88.2, 104.04]; local audio starts 0. Use attachment number and local seconds, never concatenate timestamps. Fixed gain, no retiming."
        },
        {
          "inlineData": {
            "mimeType": "audio/mpeg",
            "data": "[base64 of frozen bytes; see inputs.json]"
          }
        },
        {
          "text": "Attachment 4: Attachment 4; parent targ.mp4; EXACT parent range [80.9, 101.0]; local audio starts 0. Use attachment number and local seconds, never concatenate timestamps. Fixed gain, no retiming."
        },
        {
          "inlineData": {
            "mimeType": "audio/mpeg",
            "data": "[base64 of frozen bytes; see inputs.json]"
          }
        }
      ]
    }
  ],
  "stream": true,
  "generationConfig": {
    "thinkingConfig": {
      "includeThoughts": false,
      "thinkingLevel": "high"
    }
  }
}