CosyVoice2-API/requirements.txt at main · sin-tag/CosyVoice2-API · GitHub

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
# FastAPI and web server
fastapi>=0.104.0,<1.0.0
uvicorn[standard]>=0.24.0,<1.0.0
python-multipart>=0.0.6,<1.0.0

# WebSocket support for streaming
websockets>=11.0.0,<12.0.0

# Configuration and validation
pydantic>=2.0.0,<3.0.0
pydantic-settings>=2.0.0,<3.0.0

# Audio processing
librosa==0.10.1
soundfile==0.12.1
# PyTorch with CUDA 12.1 support
torch==2.2.2+cu121
torchaudio==2.2.2+cu121
--extra-index-url https://download.pytorch.org/whl/cu121

# Machine learning and AI
numpy<2  # Changed to avoid compatibility issues with compiled modules
scipy==1.11.4
transformers>=4.37.0,<5.0.0
modelscope==1.9.5

# Utilities
aiofiles==23.2.1
aiohttp>=3.8.0,<4.0.0
python-jose[cryptography]==3.3.0
httpx==0.25.2
tqdm==4.66.1

# CosyVoice specific dependencies
hyperpyyaml==1.2.2
onnxruntime==1.16.3
pypinyin==0.50.0
jieba==0.42.1
inflect==7.0.0
eng_to_ipa==0.0.2
unidecode==1.3.7
cn2an==0.5.22
num2words==0.5.13

# Speech processing
openai-whisper>=20231117
# WeTextProcessing>=1.0.3  # Commented out due to compilation issues
# Alternative text processing libraries
pynini  # For text normalization (WeTextProcessing dependency)
openfst-python  # OpenFST Python bindings

# Additional dependencies discovered during setup
gradio
lightning
omegaconf
diffusers
hydra-core
gdown
matplotlib
wget
pyarrow
pyworld
wetext
conformer
seaborn
wandb
tensorboard
rich
setuptools
wheel

# Development and testing
pytest==7.4.3
pytest-asyncio==0.21.1
black==23.11.0
flake8==6.1.0