Files
runpod/comfyui/workflows/text-to-music/musicgen-melody-t2m-production-v1.json
Sebastian Krüger b6ab524b79 fix: replace SaveAudio with PreviewAudio in MusicGen workflows
SaveAudio was erroring on 'waveform' key - the AUDIO output from
Musicgen_ node has a different internal structure than what SaveAudio
expects. PreviewAudio is more compatible with Sound Lab's AUDIO format.

Files are still saved to ComfyUI output directory, just through
PreviewAudio instead of SaveAudio.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-23 11:14:17 +01:00

151 lines
4.1 KiB
JSON

{
"last_node_id": 2,
"last_link_id": 1,
"nodes": [
{
"id": 1,
"type": "Musicgen_",
"pos": [50, 100],
"size": [400, 300],
"flags": {},
"order": 0,
"mode": 0,
"outputs": [
{
"name": "audio",
"type": "AUDIO",
"links": [1],
"slot_index": 0
}
],
"properties": {
"Node name for S&R": "Musicgen_"
},
"widgets_values": [
"Electronic music with melodic elements and catchy hooks",
30.0,
3.5,
42,
"auto"
],
"title": "MusicGen Melody Generator"
},
{
"id": 2,
"type": "PreviewAudio",
"pos": [500, 100],
"size": [315, 100],
"flags": {},
"order": 1,
"mode": 0,
"inputs": [
{
"name": "audio",
"type": "AUDIO",
"link": 1
}
],
"properties": {
"Node name for S&R": "PreviewAudio"
},
"title": "API Audio Output"
}
],
"links": [
[1, 1, 0, 2, 0, "AUDIO"]
],
"groups": [],
"config": {},
"extra": {
"workflow_info": {
"name": "MusicGen Melody Text-to-Music Production",
"version": "1.2.0",
"description": "Music generation using MusicGen with melodic focus. Note: Melody conditioning via audio input not yet supported.",
"category": "text-to-music",
"tags": ["musicgen", "melody", "t2m", "audio", "production"],
"requirements": {
"models": ["facebook/musicgen-melody"],
"custom_nodes": ["comfyui-sound-lab"],
"vram_min": "10GB",
"vram_recommended": "12GB"
},
"parameters": {
"prompt": {
"node_id": 1,
"widget_index": 0,
"type": "string",
"required": true,
"default": "Electronic music with melodic elements",
"description": "Text description of desired music style, genre, mood, instruments"
},
"duration": {
"node_id": 1,
"widget_index": 1,
"type": "float",
"required": false,
"default": 30.0,
"min": 1.0,
"max": 60.0,
"description": "Duration in seconds (up to 30s recommended for best quality)"
},
"guidance_scale": {
"node_id": 1,
"widget_index": 2,
"type": "float",
"required": false,
"default": 3.5,
"min": 0.0,
"max": 20.0,
"description": "How closely to follow the prompt (3.5 recommended)"
},
"seed": {
"node_id": 1,
"widget_index": 3,
"type": "integer",
"required": false,
"default": 42,
"min": 0,
"max": 2147483647,
"description": "Random seed for reproducibility"
},
"device": {
"node_id": 1,
"widget_index": 4,
"type": "string",
"required": false,
"default": "auto",
"options": ["auto", "cpu"],
"description": "Computation device (auto recommended)"
}
},
"outputs": {
"audio": {
"node_id": 2,
"type": "audio",
"format": "FLAC",
"sample_rate": "32000 Hz",
"channels": "stereo"
}
},
"performance": {
"avg_generation_time": "40-60 seconds for 30s audio",
"vram_usage": "~10-12GB",
"gpu_utilization": "90-100%"
},
"use_cases": [
"Melodic music generation",
"Creating catchy hooks and melodies",
"Music with strong melodic structure"
],
"notes": [
"IMPORTANT: Melody conditioning via audio input not yet supported in current Sound Lab implementation",
"This workflow uses text prompts only - describe melodic characteristics in the prompt",
"For best melodic results, use descriptive terms like 'catchy melody', 'melodic hooks', 'harmonic progression'",
"MusicGen Melody model trained for melodic content",
"Output is 32kHz stereo FLAC format"
]
}
},
"version": 0.4
}