Bug: CPU loading fails when device=torch.device("cpu")

Environment:

• Windows 11
• Python 3.11
• CPU-only machine

Code:

from chatterbox.mtl_tts import ChatterboxMultilingualTTS import torch

model = ChatterboxMultilingualTTS.from_pretrained( device=torch.device("cpu") )

Result:

RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False

Cause:

In D:\chatterbox\Lib\site-packages\chatterbox\mtl_tts.py In from_local():

if device in ["cpu", "mps"]: map_location = torch.device("cpu") else: map_location = None

When device is torch.device("cpu"), the condition evaluates to False.

Fix Instead of if device in ["cpu", "mps"]: map_location = torch.device("cpu") else: map_location = None

replace:

if str(device) in ["cpu", "mps"]: map_location = torch.device("cpu") else: map_location = None