Update app.py

app.py

@@ -3,33 +3,33 @@ from transformers import WhisperProcessor, WhisperForConditionalGeneration
 from datasets import load_dataset
 import torch
 
-# Load Whisper model and processor
+# 加载 Whisper 模型和 processor
 model_name = "openai/whisper-large-v3-turbo"
 processor = WhisperProcessor.from_pretrained(model_name)
 model = WhisperForConditionalGeneration.from_pretrained(model_name)
 
-# Load dataset (bigcode/the-stack)
-dataset = load_dataset("bigcode/the-stack", data_dir="data/html")
+# 加载数据集 bigcode/the-stack
+dataset = load_dataset("bigcode/the-stack", data_dir="data/html", split="train")
 
 def transcribe(audio):
-    # Process audio for transcription
+    # 处理音频进行转录
     audio_input = processor(audio, return_tensors="pt").input_values
     with torch.no_grad():
         logits = model(audio_input).logits
     predicted_ids = torch.argmax(logits, dim=-1)
     transcription = processor.batch_decode(predicted_ids)
     
-    # Return the transcription
+    # 返回转录结果
     return transcription[0]
 
-# Gradio interface
+# Gradio 界面
 iface = gr.Interface(
     fn=transcribe,
     inputs=gr.Audio(source="microphone", type="filepath"),
     outputs="text",
     title="Whisper Transcription for Developers",
-    description="Transcribe developer-related terms using Whisper and bigcode dataset for contextual support."
+    description="使用 Whisper 和 bigcode 数据集转录开发者相关术语。"
 )
 
-# Launch the Gradio app
+# 启动 Gradio 应用
 iface.launch()