remove unnecessary code

cosyvoice/flow/flow.py

@@ -113,7 +113,7 @@ class MaskedDiffWithXvec(torch.nn.Module):
         # concat text and prompt_text
         token_len1, token_len2 = prompt_token.shape[1], token.shape[1]
         token, token_len = torch.concat([prompt_token, token], dim=1), prompt_token_len + token_len
-        mask = (~make_pad_mask(token_len)).to(embedding.dtype).unsqueeze(-1).to(embedding)
+        mask = (~make_pad_mask(token_len)).unsqueeze(-1).to(embedding)
         token = self.input_embedding(torch.clamp(token, min=0)) * mask
 
         # text encode

cosyvoice/flow/flow_matching.py

@@ -14,8 +14,6 @@
 import torch
 import torch.nn.functional as F
 from matcha.models.components.flow_matching import BASECFM
-import onnxruntime as ort
-import numpy as np
 
 class ConditionalCFM(BASECFM):
     def __init__(self, in_channels, cfm_params, n_spks=1, spk_emb_dim=64, estimator: torch.nn.Module = None):