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
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
|
"""Configuration for the audiobook converter.
Edit these values to change the default voice and processing behavior.
All paths are resolved relative to the project root, so the converter can
be run from any working directory.
"""
from pathlib import Path
# Project root (directory containing audiobook.py)
BASE_DIR = Path(__file__).resolve().parent.parent
# =============================================================================
# VOICE MODES
# =============================================================================
VOICE_MODE_CUSTOM = "custom_voice"
VOICE_MODE_CLONE = "voice_clone"
VOICE_MODES = (VOICE_MODE_CUSTOM, VOICE_MODE_CLONE)
# =============================================================================
# TTS LANGUAGE SETTINGS
# =============================================================================
# Languages understood by the Qwen3-TTS API. The Gradio demo silently falls
# back to "Auto" for unrecognized values, so languages are validated
# client-side (see converter.tts.normalize_language) before reaching the API.
# Display names must match the demo dropdown exactly.
TTS_LANGUAGES = (
"Auto",
"Chinese",
"English",
"German",
"Italian",
"Portuguese",
"Spanish",
"Japanese",
"Korean",
"French",
"Russian",
)
# Short aliases accepted on the command line (ISO 639-1 codes and common
# shorthands), mapped to the display names above.
TTS_LANGUAGE_ALIASES = {
"zh": "Chinese",
"en": "English",
"de": "German",
"it": "Italian",
"pt": "Portuguese",
"es": "Spanish",
"ja": "Japanese",
"ko": "Korean",
"fr": "French",
"ru": "Russian",
"zh-cn": "Chinese",
"zh-tw": "Chinese",
"pt-br": "Portuguese",
"en-us": "English",
"en-gb": "English",
}
# =============================================================================
# QWEN API CONFIGURATION
# =============================================================================
QWEN_API_URL = "http://127.0.0.1:7860" # CustomVoice demo endpoint
API_TIMEOUT = 300 # Seconds before an API call times out
MAX_RETRIES = 3 # Retry failed chunks
# =============================================================================
# CUSTOM VOICE SETTINGS (pre-built speakers, always uses the 1.7B model)
# =============================================================================
CUSTOM_VOICE_SPEAKER = "Vivian"
CUSTOM_VOICE_LANGUAGE = "English"
CUSTOM_VOICE_INSTRUCT = "Speak naturally and clearly, as if reading a dramatic book to an adult audience."
CUSTOM_VOICE_MODEL_SIZE = "1.7B"
CUSTOM_VOICE_SEED = -1
CUSTOM_VOICE_MODEL_ID = "Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice"
# Map canonical speaker names to the display names used by the qwen-tts Gradio demo.
SPEAKER_DISPLAY_NAMES = {
"ryan": "Ryan",
"serena": "Serena",
"vivian": "Vivian",
"uncle_fu": "Uncle Fu",
"aiden": "Aiden",
"ono_anna": "Ono Anna",
"sohee": "Sohee",
"eric": "Eric",
"dylan": "Dylan",
}
# =============================================================================
# VOICE CLONE SETTINGS (clone a voice from a reference audio file)
# =============================================================================
# Voice clone requires the Base-model demo (Qwen3-TTS-12Hz-1.7B-Base), which
# exposes /run_voice_clone. The CustomVoice demo only exposes /run_instruct, so
# run the Base demo on a separate port and point this at it.
VOICE_CLONE_LANGUAGE = "English"
VOICE_CLONE_USE_XVECTOR_ONLY = False
VOICE_CLONE_MODEL_SIZE = "1.7B"
VOICE_CLONE_MAX_CHUNK_CHARS = 200
VOICE_CLONE_CHUNK_GAP = 0
VOICE_CLONE_SEED = -1
VOICE_CLONE_API_URL = "http://127.0.0.1:7861"
# =============================================================================
# PROCESSING SETTINGS
# =============================================================================
BOOKS_FOLDER = BASE_DIR / "input" # Input folder
AUDIOBOOKS_FOLDER = BASE_DIR / "output" # Output folder
CHUNKS_FOLDER = BASE_DIR / "chunks" # Scratch space for per-chunk audio (cleaned per book)
LOGS_FOLDER = BASE_DIR / "logs"
CHUNK_SIZE_WORDS = 1500 # Words per TTS chunk
MIN_DELAY_BETWEEN_CHUNKS = 1 # Seconds between API calls
HEARTBEAT_INTERVAL_SECONDS = 30 # Print "still working" this often during a chunk
# =============================================================================
# AUDIO OUTPUT SETTINGS
# =============================================================================
AUDIO_FORMAT = "m4b" # Default output container
AUDIO_FORMATS = ("mp3", "m4b", "ogg", "flac")
AUDIO_BITRATE = "128k"
SUPPORTED_FORMATS = [".txt", ".pdf", ".epub"]
|
