Spaces:
Sleeping
Sleeping
File size: 1,764 Bytes
90c1221 |
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 |
import sys
import torch
from transformers import ClapModel, ClapProcessor
from config import config
models = dict()
LOCAL_PATH = "./emotional/clap-htsat-fused"
processor = ClapProcessor.from_pretrained(LOCAL_PATH)
def get_clap_audio_feature(audio_data, device=config.bert_gen_config.device):
if (
sys.platform == "darwin"
and torch.backends.mps.is_available()
and device == "cpu"
):
device = "mps"
if not device:
device = "cuda"
if device not in models.keys():
if config.webui_config.fp16_run:
models[device] = ClapModel.from_pretrained(
LOCAL_PATH, torch_dtype=torch.float16
).to(device)
else:
models[device] = ClapModel.from_pretrained(LOCAL_PATH).to(device)
with torch.no_grad():
inputs = processor(
audios=audio_data, return_tensors="pt", sampling_rate=48000
).to(device)
emb = models[device].get_audio_features(**inputs).float()
return emb.T
def get_clap_text_feature(text, device=config.bert_gen_config.device):
if (
sys.platform == "darwin"
and torch.backends.mps.is_available()
and device == "cpu"
):
device = "mps"
if not device:
device = "cuda"
if device not in models.keys():
if config.webui_config.fp16_run:
models[device] = ClapModel.from_pretrained(
LOCAL_PATH, torch_dtype=torch.float16
).to(device)
else:
models[device] = ClapModel.from_pretrained(LOCAL_PATH).to(device)
with torch.no_grad():
inputs = processor(text=text, return_tensors="pt").to(device)
emb = models[device].get_text_features(**inputs).float()
return emb.T
|