fixed bugs with hf tokens

autotranscript/cli.py

@@ -8,6 +8,8 @@ import os
 from argparse import ArgumentParser, ArgumentDefaultsHelpFormatter
 import json
 
+from sympy import use
+
 from .autotranscript import AutoTranscribe
 
 from whisper.tokenizer import LANGUAGES , TO_LANGUAGE_CODE
@@ -57,12 +59,9 @@ def cli():
     parser.add_argument("--diarization_directory", type=str, default= None,
                         help="Path to the diarization model directory.")
 
-    parser.add_argument("--huggingface_token", default= None, type=str,
+    parser.add_argument("--hf_token", default= None, type=str,
                         help="HuggingFace token for private model download.")
 
-    parser.add_argument("--allow_download", type=str2bool, default=True,
-                        help="Allow model download if not found locally.")
-
     parser.add_argument("--inference_device",
                         default="cuda" if is_available() else "cpu",
                         help="Device to use for PyTorch inference.")
@@ -108,12 +107,12 @@ def cli():
     if args.num_threads > 0:
         set_num_threads(arg_dict.pop("num_threads"))
     
-    class_kwargs = dict()
+    class_kwargs = {'whisper_model' : arg_dict.pop("whisper_model_name"),
-    
+                    'dia_model': arg_dict.pop("diarization_directory"),
-    for k, v in arg_dict.items():
+                    'use_auth_token' : arg_dict.pop("hf_token")}
-        if v is not None:
-            class_kwargs[k] = v
     
+    if arg_dict["whisper_model_directory"]:
+        class_kwargs["download_root"] = arg_dict.pop("whisper_model_directory")
 
     model = AutoTranscribe(**class_kwargs)
 

autotranscript/diarisation.py

@@ -208,6 +208,8 @@ class Diariser:
         if not os.path.exists(model) and use_auth_token is None:
             use_auth_token = cls._get_token()
             model = 'pyannote/speaker-diarization'
+        elif not os.path.exists(model) and use_auth_token is not None:
+            model = 'pyannote/speaker-diarization'
             
         _model =  Pipeline.from_pretrained(model,
                                            use_auth_token = use_auth_token,