Update app.py

app.py

@@ -10,19 +10,24 @@ model = WhisperForConditionalGeneration.from_pretrained(model_name)
 
 # 加载数据集 bigcode/the-stack
 
-# ds = load_dataset("CoIR-Retrieval/CodeSearchNet-php-queries-corpus")
+ds = load_dataset("CoIR-Retrieval/CodeSearchNet-php-queries-corpus")
 
-def transcribe(audio):
-    # 处理音频进行转录
-    audio_input = processor(audio, return_tensors="pt").input_values
+def transcribe(audio_path):
+    # 加载音频文件并转换为信号
+    audio, sr = librosa.load(audio_path, sr=16000)
+    input_values = processor(audio, return_tensors="pt", sampling_rate=16000).input_values
+
+    # 模型推理
     with torch.no_grad():
-        logits = model(audio_input).logits
+        logits = model(input_values).logits
+
     predicted_ids = torch.argmax(logits, dim=-1)
     transcription = processor.batch_decode(predicted_ids)
     
     # 返回转录结果
     return transcription[0]
 
+    
 # Gradio 界面
 iface = gr.Interface(
     fn=transcribe,