(wip)debug

tts.py

@@ -47,7 +47,7 @@ data = {"text": "string", "provider": "string", "model": "string"}
 
 def predict_index_tts(text, reference_audio_path=None):
     from gradio_client import Client, handle_file
-    client = Client("kemuriririn/IndexTTS", hf_token=get_zerogpu_token())
+    client = Client("kemuriririn/IndexTTS")
     if reference_audio_path:
         prompt = handle_file(reference_audio_path)
     else:
@@ -65,7 +65,7 @@ def predict_index_tts(text, reference_audio_path=None):
 
 def predict_spark_tts(text, reference_audio_path=None):
     from gradio_client import Client, handle_file
-    client = Client("kemuriririn/SparkTTS", hf_token=get_zerogpu_token())
+    client = Client("kemuriririn/SparkTTS")
     prompt_wav = None
     if reference_audio_path:
         prompt_wav = handle_file(reference_audio_path)
@@ -82,7 +82,7 @@ def predict_spark_tts(text, reference_audio_path=None):
 
 def predict_cosyvoice_tts(text, reference_audio_path=None):
     from gradio_client import Client, file, handle_file
-    client = Client("kemuriririn/CosyVoice2-0.5B", hf_token=os.getenv("HF_TOKEN"))
+    client = Client("kemuriririn/CosyVoice2-0.5B")
     if not reference_audio_path:
         raise ValueError("cosyvoice-2.0 需要 reference_audio_path")
     prompt_wav = handle_file(reference_audio_path)
@@ -125,7 +125,7 @@ def predict_maskgct(text, reference_audio_path=None):
 
 def predict_gpt_sovits_v2(text, reference_audio_path=None):
     from gradio_client import Client, file
-    client = Client("kemuriririn/GPT-SoVITS-v2", hf_token=os.getenv("HF_TOKEN"))
+    client = Client("kemuriririn/GPT-SoVITS-v2")
     if not reference_audio_path:
         raise ValueError("GPT-SoVITS-v2 需要 reference_audio_path")
     result = client.predict(