Nodes/ComfyUI-MiniMaxH3Audio/MiniMax H3 Audio Prompt
ComfyUI Node

MiniMax H3 Audio Prompt

H3 T2VA 形式(integrated_multimodal_description / overall_soundscape / non_diegetic_music)で音楽プロンプトを合成します。

By t22m003·Created 14 days ago·Updated 14 days ago· 0
MiniMax H3 Audio Prompt
    • prompt
    non_diegetic_musicwarm lo-fi hip-hop beat at 86 BPM, dusty vinyl crackle, mellow electric piano chords, deep rounded bass, soft snare brushes, gentle swing groove, stereo width, no vocals
    overall_soundscapeN/A
    sceneCinematic night city skyline with soft neon reflections on wet asphalt; no people on screen; the camera holds a slow static wide shot while the soundtrack carries the emotion.
    seconds15.0
    Categoryaudio/minimax_h3

    Inputs (4)

    NameTypeDefaultDescription
    non_diegetic_musicSTRINGwarm lo-fi hip-hop beat at 86 BPM, dusty vinyl crackle, mellow electric piano chords, deep rounded bass, soft snare brushes, gentle swing groove, stereo width, no vocals楽器・テンポ・ダイナミクス(観客だけが聴く BGM)
    overall_soundscapeSTRINGN/A環境音の要約。無ければ N/A
    sceneSTRINGCinematic night city skyline with soft neon reflections on wet asphalt; no people on screen; the camera holds a slow static wide shot while the soundtrack carries the emotion.integrated_multimodal_description に入れる情景(黒画面より情景込みが有利な場合あり)
    secondsFLOAT15.05.16–150尺ヒント(プロンプト文言にのみ反映。実際の latent 尺は TextToAudio.seconds)

    Outputs (1)

    NameTypeDescription
    promptSTRING