Hugging Face's logo

Spaces:
ACE-Step
/
Ace-Step-v1.5
Running on A100

App Files Community
8
Ace-Step-v1.5
File size: 4,153 Bytes 
1241c80
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4477394
e4e90cd
 
1241c80
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1da0418
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
 
"""
Constants for ACE-Step
Centralized constants used across the codebase
"""

# ==============================================================================
# Language Constants
# ==============================================================================

VALID_LANGUAGES = [
    'ar', 'az', 'bg', 'bn', 'ca', 'cs', 'da', 'de', 'el', 'en',
    'es', 'fa', 'fi', 'fr', 'he', 'hi', 'hr', 'ht', 'hu', 'id',
    'is', 'it', 'ja', 'ko', 'la', 'lt', 'ms', 'ne', 'nl', 'no',
    'pa', 'pl', 'pt', 'ro', 'ru', 'sa', 'sk', 'sr', 'sv', 'sw',
    'ta', 'te', 'th', 'tl', 'tr', 'uk', 'ur', 'vi', 'yue', 'zh',
    'unknown'
]


# ==============================================================================
# Keyscale Constants
# ==============================================================================

KEYSCALE_NOTES = ['A', 'B', 'C', 'D', 'E', 'F', 'G']
KEYSCALE_ACCIDENTALS = ['', '#', 'b', '♯', '♭']  # empty + ASCII sharp/flat + Unicode sharp/flat
KEYSCALE_MODES = ['major', 'minor']

# Generate all valid keyscales: 7 notes × 5 accidentals × 2 modes = 70 combinations
VALID_KEYSCALES = set()
for note in KEYSCALE_NOTES:
    for acc in KEYSCALE_ACCIDENTALS:
        for mode in KEYSCALE_MODES:
            VALID_KEYSCALES.add(f"{note}{acc} {mode}")


# ==============================================================================
# Metadata Range Constants
# ==============================================================================

# BPM (Beats Per Minute) range
BPM_MIN = 30
BPM_MAX = 300

# Duration range (in seconds)
DURATION_MIN = 10
DURATION_MAX = 600

# Valid time signatures
VALID_TIME_SIGNATURES = [2, 3, 4, 6]


# ==============================================================================
# Task Type Constants
# ==============================================================================

TASK_TYPES = ["text2music", "repaint", "cover", "extract", "lego", "complete"]

# Task types available for turbo models (subset)
TASK_TYPES_TURBO = ["text2music", "repaint", "cover"]

# Task types available for base models (full set)
TASK_TYPES_BASE = ["text2music", "repaint", "cover", "extract", "lego", "complete"]


# ==============================================================================
# Instruction Constants
# ==============================================================================

# Default instructions
DEFAULT_DIT_INSTRUCTION = "Fill the audio semantic mask based on the given conditions:"
DEFAULT_LM_INSTRUCTION = "Generate audio semantic tokens based on the given conditions:"
DEFAULT_LM_UNDERSTAND_INSTRUCTION = "Understand the given musical conditions and describe the audio semantics accordingly:"
DEFAULT_LM_INSPIRED_INSTRUCTION = "Expand the user's input into a more detailed and specific musical description:"
DEFAULT_LM_REWRITE_INSTRUCTION = "Format the user's input into a more detailed and specific musical description:"

# Instruction templates for each task type
# Note: Some instructions use placeholders like {TRACK_NAME} or {TRACK_CLASSES}
# These should be formatted using .format() or f-strings when used
TASK_INSTRUCTIONS = {
    "text2music": "Fill the audio semantic mask based on the given conditions:",
    "repaint": "Repaint the mask area based on the given conditions:",
    "cover": "Generate audio semantic tokens based on the given conditions:",
    "extract": "Extract the {TRACK_NAME} track from the audio:",
    "extract_default": "Extract the track from the audio:",
    "lego": "Generate the {TRACK_NAME} track based on the audio context:",
    "lego_default": "Generate the track based on the audio context:",
    "complete": "Complete the input track with {TRACK_CLASSES}:",
    "complete_default": "Complete the input track:",
}


# ==============================================================================
# Track/Instrument Constants
# ==============================================================================

TRACK_NAMES = [
    "woodwinds", "brass", "fx", "synth", "strings", "percussion",
    "keyboard", "guitar", "bass", "drums", "backing_vocals", "vocals"
]

SFT_GEN_PROMPT = """# Instruction
{}

# Caption
{}

# Metas
{}<|endoftext|>
"""