add api_token

app.py

@@ -3,7 +3,7 @@ import torch
 from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
 import gradio as gr
 import sox
-
+import os
 
 def convert(inputfile, outfile):
     sox_tfm = sox.Transformer()
@@ -11,11 +11,12 @@ def convert(inputfile, outfile):
         file_type="wav", channels=1, encoding="signed-integer", rate=16000, bits=16
     )
     sox_tfm.build(inputfile, outfile)
-    
 
+
+api_token = os.getenv("API_TOKEN")
 model_name = "indonesian-nlp/wav2vec2-indonesian-javanese-sundanese"
-processor = Wav2Vec2Processor.from_pretrained(model_name)
-model = Wav2Vec2ForCTC.from_pretrained(model_name)
+processor = Wav2Vec2Processor.from_pretrained(model_name, use_auth_token=api_token)
+model = Wav2Vec2ForCTC.from_pretrained(model_name, use_auth_token=api_token)
 
 def parse_transcription(wav_file):
     filename = wav_file.name.split('.')[0]
@@ -30,16 +31,16 @@ def parse_transcription(wav_file):
 
 
 output = gr.outputs.Textbox(label="Indonesian, Javanese or Sundanese")
-    
-input_ = gr.inputs.Audio(source="microphone", type="file") 
-#gr.Interface(parse_transcription, inputs = input_,  outputs="text", 
-#             analytics_enabled=False, show_tips=False, enable_queue=True).launch(inline=False);
-             
-gr.Interface(parse_transcription, inputs = input_,  outputs=[output],
+
+input_ = gr.inputs.Audio(source="microphone", type="file")
+
+gr.Interface(parse_transcription, inputs=input_,  outputs=[output],
              analytics_enabled=False,
              show_tips=False,
              theme='huggingface',
              layout='vertical',
              title="Multilingual Speech Recognition for Indonesian Languages",
-             description="Speech Recognition Live Demo for Indonesian, Javanese and Sundanese",
+             description="This is Speech Recognition live demo for Indonesian, Javanese and Sundanese Language. It uses"
+                         "the Wav2Vec2 large model \"indonesian-nlp/wav2vec2-indonesian-javanese-sundanese\""
+                         " fine-tuned on Common Voice and OpenSLR speech datasets.",
              enable_queue=True).launch( inline=False)