diff --git a/pretrain/meta.py b/pretrain/meta.py index cc35dd3..c591573 100644 --- a/pretrain/meta.py +++ b/pretrain/meta.py @@ -12,9 +12,17 @@ def download_dict(): "url": "https://github.com/bshall/hubert/releases/download/v0.1/hubert-soft-0d54a1f4.pt", "output": "./pretrain/hubert-soft-0d54a1f4.pt" }, + "whisper-ppg-small": { + "url": "https://openaipublic.azureedge.net/main/whisper/models/9ecf779972d90ba49c06d968637d720dd632c55bbf19d441fb42bf17a411e794/small.pt", + "output": "./pretrain/small.pt" + }, "whisper-ppg": { "url": "https://openaipublic.azureedge.net/main/whisper/models/345ae4da62f9b3d59415adc60127b97c714f32e89e936602e85993674d08dcb1/medium.pt", "output": "./pretrain/medium.pt" + }, + "whisper-ppg-large": { + "url": "https://openaipublic.azureedge.net/main/whisper/models/81f7c96c852ee8fc832187b0132e569d6c3065a3252ed18e56effd0b6a73e524/large-v2.pt", + "output": "./pretrain/large-v2.pt" } } diff --git a/requirements.txt b/requirements.txt index 8e22d3e..d09b883 100644 --- a/requirements.txt +++ b/requirements.txt @@ -23,5 +23,4 @@ transformers edge_tts pyyaml pynvml -ffmpeg faiss-cpu diff --git a/requirements_win.txt b/requirements_win.txt index 19bd0e3..e35a80e 100644 --- a/requirements_win.txt +++ b/requirements_win.txt @@ -26,5 +26,4 @@ transformers edge_tts pyyaml pynvml -ffmpeg -faiss \ No newline at end of file +faiss-cpu diff --git a/sovits4_for_colab.ipynb b/sovits4_for_colab.ipynb index 045f8f6..2dcf532 100644 --- a/sovits4_for_colab.ipynb +++ b/sovits4_for_colab.ipynb @@ -280,7 +280,7 @@ "\n", "%cd /content/so-vits-svc\n", "\n", - "speech_encoder = \"vec768l12\" #@param [\"vec768l12\", \"vec256l9\", \"hubertsoft\", \"whisper-ppg\"]\n", + "speech_encoder = \"vec768l12\" #@param [\"vec768l12\", \"vec256l9\", \"hubertsoft\", \"whisper-ppg\", \"whisper-ppg-large\"]\n", "\n", "from pretrain.meta import download_dict\n", "download_dict = download_dict()\n",