VoiceCraft/cog.yaml

26 lines
1.2 KiB
YAML
Raw Normal View History

2024-04-05 15:20:11 +00:00
# Configuration for Cog ⚙️
# Reference: https://github.com/replicate/cog/blob/main/docs/yaml.md
build:
gpu: true
system_packages:
2024-04-19 10:46:00 +00:00
- libgl1-mesa-glx
- libglib2.0-0
2024-04-05 15:20:11 +00:00
- ffmpeg
- espeak-ng
2024-04-19 10:46:00 +00:00
python_version: "3.11"
2024-04-05 15:20:11 +00:00
python_packages:
2024-04-19 10:46:00 +00:00
- torch==2.1.0
- torchaudio==2.1.0
- xformers
2024-04-05 15:20:11 +00:00
- phonemizer==3.2.1
2024-04-14 12:15:23 +00:00
- whisperx==3.1.1
- openai-whisper>=20231117
2024-04-05 15:20:11 +00:00
run:
2024-04-19 10:48:15 +00:00
# - git clone https://github.com/facebookresearch/audiocraft && pip install -e ./audiocraft
2024-04-19 10:49:04 +00:00
- pip install -e git+https://github.com/facebookresearch/audiocraft.git@c5157b5bf14bf83449c17ea1eeb66c19fb4bc7f0#egg=audiocraft # use "git clone https://github.com/facebookresearch/audiocraft && pip install -e ./audiocraft" instead if hits audiocraft import error
2024-04-05 15:20:11 +00:00
- pip install "pydantic<2.0.0"
- curl -o /usr/local/bin/pget -L "https://github.com/replicate/pget/releases/download/v0.6.0/pget_linux_x86_64" && chmod +x /usr/local/bin/pget
2024-04-19 10:46:00 +00:00
- mkdir -p /root/.cache/torch/hub/checkpoints/ && wget --output-document "/root/.cache/torch/hub/checkpoints/wav2vec2_fairseq_base_ls960_asr_ls960.pth" "https://download.pytorch.org/torchaudio/models/wav2vec2_fairseq_base_ls960_asr_ls960.pth"
predict: "predict.py:Predictor"