Update app.py

app.py

@@ -64,17 +64,64 @@ dict_frog_desc={'American_bull':'赤蛙科，是一種水生青蛙。是現存
 
 loaded_model = joblib.load('./RFC_model')
 
+def Audio(fname,SAMPLE_RATE):
+    y, sr = librosa.load(fname, sr=SAMPLE_RATE)    
+    return y, sr
+
+def AudioImg(y,sr):      
+    img=librosa.display.waveplot(y, sr)
+    fig = plt.figure()
+    plot_img_np = get_img_from_fig(fig) 
+    return plot_img_np
+
+def MFCC_spectrogram(y,sr):
+    # 提取頻域音頻資料 MFCC spectrogram feature  
+    melspec = librosa.feature.melspectrogram(y, sr, n_fft=1024, hop_length=512, n_mels=128)  
+    # 轉成 log scale 已呈現頻譜資料  
+    logmelspec = librosa.power_to_db(melspec)  
+    # 繪圖    
+    img=librosa.display.specshow(logmelspec, sr=sr, x_axis='time', y_axis='mel')
+    fig = plt.figure()
+    plot_img_np = get_img_from_fig(fig)
+    #plt.title('Features of unknow frog by MFCC (Mel Frquency Cepstral Coefficients)')  
+    #plt.show()
+    return plot_img_np
+    
+def get_img_from_fig(fig, dpi=180):
+    buf = io.BytesIO()
+    fig.savefig(buf, format="png", dpi=dpi)
+    buf.seek(0)
+    img_arr = np.frombuffer(buf.getvalue(), dtype=np.uint8)
+    buf.close()
+    img = cv2.imdecode(img_arr, 1)
+    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+    return img
+
+
 def flog(A):
     test_scaled = FeatureExtraction(A)
     result = loaded_model.predict(test_scaled)
     img=cv2.imread('./'+labels[result[0]]+'.jpg')
     img=cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+    
+    y,sr=Audio(A,SAMPLE_RATE)
+    fig = plt.figure()
+    plt.title("waveplot",fontsize=20)
+    AudioImg(y,sr)
+    img2 = get_img_from_fig(fig)
+
+    fig = plt.figure()
+    plt.title("MFCC Feature",fontsize=20)
+    MFCC_spectrogram(y,sr)
+    img3 = get_img_from_fig(fig)
     #return dict_frog[labels[result[0]]],img
-    return dict_frog[labels[result[0]]],dict_frog_desc[labels[result[0]]],img
+    return dict_frog[labels[result[0]]],dict_frog_desc[labels[result[0]]],img,img2,img3
 
 output_label = gr.Label()
 output_desc = gr.Label()
 output_image = gr.Image()
+output_image2 = gr.Image()
+output_image3 = gr.Image()
 
 title="台灣青蛙聲音辨識"
 
@@ -104,6 +151,6 @@ gr.Interface(flog,
   inputs=[
     gr.inputs.Audio(source="microphone", type="filepath"),
   ],
-  outputs=[output_label,output_desc,output_image],title=title,description=description).launch()
+  outputs=[output_label,output_desc,output_image,output_image2,output_image3],title=title,description=description).launch()
   #outputs=[output_label,output_image],title=title,description=description).launch()