change Arthur.config to lower standarad

app.py

@@ -19,7 +19,7 @@ from transformers import pipeline
 import librosa
 import librosa.display
 import matplotlib.pyplot as plt
-
+from local.convert_metrics import nat2avaMOS, WER2INTELI
 
 # Google cloud service
 from googleapiclient.discovery import build
@@ -229,8 +229,14 @@ def calc_mos(_, audio_path, id, ref, pre_ppm, fig=None):
             "GOOD JOB! Please 【Save the Recording】.\nYou can start recording the next sample."
         )
 
-    # Google Drive saving # TODO
-    click_google_saving(audio_path)
+    # Google Drive saving 
+    saved_google_id = None
+    if error_msg == ("GOOD JOB! Please 【Save the Recording】.\nYou can start recording the next sample."):
+        saved_google_id = click_google_saving(audio_path)
+        # TODO: add saved_google_id to the csv file
+    ## else:
+    ## TODO: clear all output as start recording again
+    ##     print("Saving Failed")
 
     return (
         fig_h,
@@ -240,6 +246,7 @@ def calc_mos(_, audio_path, id, ref, pre_ppm, fig=None):
         phone_transcription,
         ppm,
         error_msg,
+        saved_google_id,
     )
 
 with open("src/description.html", "r", encoding="utf-8") as f:
@@ -345,7 +352,6 @@ def generate_now_time_wav():
 
 # Add google drive cloud saving
 def click_google_saving(audio_file,
-    
                         ):
                         # reference_id,
                         # reference_textbox,
@@ -361,8 +367,7 @@ def click_google_saving(audio_file,
     
     request = service.files().create(
         media_body=media,
-        body={'name': name, 
-              }
+        body={'name': name, }
     )
             #   'reference_id': reference_id,
             #   "reference_textbox": reference_textbox,
@@ -373,6 +378,8 @@ def click_google_saving(audio_file,
             #   "ppm": ppm, 
             #   "msg": msg
     response = request.execute()
+    # get saved file id
+    return response.get('id')
     # return response.get('id')
     
 
@@ -404,18 +411,19 @@ with gr.Blocks(css=css, theme=my_theme) as demo:
                     interactive=True,
                     elem_classes="ref_text",
                 )
-                with gr.Accordion("Input for Development", open=False):
-                    reference_id = gr.Textbox(
-                        value="ID",
-                        placeholder="Utter ID",
-                        label="Reference_ID",
-                        visible=True,
-                    )
-                    reference_PPM = gr.Textbox(
-                        placeholder="Pneumatic Voice's PPM",
-                        label="Ref PPM",
-                        visible=True,
-                    )
+        with gr.Row():
+            with gr.Accordion("Input for Development", open=False):
+                reference_id = gr.Textbox(
+                    value="ID",
+                    placeholder="Utter ID",
+                    label="Reference_ID",
+                    visible=True,
+                )
+                reference_PPM = gr.Textbox(
+                    placeholder="Pneumatic Voice's PPM",
+                    label="Ref PPM",
+                    visible=True,
+                )
         with gr.Row():
             b = gr.Button(value="1.Submit", variant="primary", elem_classes="submit")
 
@@ -440,42 +448,48 @@ with gr.Blocks(css=css, theme=my_theme) as demo:
                 interactive=False,
                 elem_classes="message",
             )
-            with gr.Accordion("Output for Development", open=False):
-                wav_plot = gr.Plot(PlaceHolder="Wav/Pause Plot", label="wav_pause_plot", visible=True)
+        with gr.Accordion("Output for Development", open=False):
+            wav_plot = gr.Plot(PlaceHolder="Wav/Pause Plot", label="wav_pause_plot", visible=True)
 
-                predict_mos = gr.Textbox(
-                    placeholder="Predicted MOS",
-                    label="Predicted MOS",
-                    visible=True,
-                )
+            predict_mos = gr.Textbox(
+                placeholder="Predicted MOS",
+                label="Predicted MOS",
+                visible=True,
+            )
 
-                hyp = gr.Textbox(placeholder="Hypothesis", label="Hypothesis", visible=True)
+            hyp = gr.Textbox(placeholder="Hypothesis", label="Hypothesis", visible=True)
 
-                wer = gr.Textbox(placeholder="Word Error Rate", label="WER", visible=True)
+            wer = gr.Textbox(placeholder="Word Error Rate", label="WER", visible=True)
 
-                predict_pho = gr.Textbox(
-                    placeholder="Predicted Phonemes",
-                    label="Predicted Phonemes",
-                    visible=True,
-                )
+            predict_pho = gr.Textbox(
+                placeholder="Predicted Phonemes",
+                label="Predicted Phonemes",
+                visible=True,
+            )
 
-                ppm = gr.Textbox(
-                    placeholder="Phonemes per minutes",
-                    label="PPM",
-                    visible=True,
-                )
-            outputs = [
-                wav_plot,
-                predict_mos,
-                hyp,
-                wer,
-                predict_pho,
-                ppm,
-                msg,
-            ]
+            ppm = gr.Textbox(
+                placeholder="Phonemes per minutes",
+                label="PPM",
+                visible=True,
+            )
+            saved_google_drive_id = gr.Textbox(
+                placeholder="Saved Google Drive ID",
+                label="Saved Google Drive ID",
+                visible=True,
+            )
+        outputs = [
+            wav_plot,
+            predict_mos,
+            hyp,
+            wer,
+            predict_pho,
+            ppm,
+            msg,
+            saved_google_drive_id
+        ]
 
-            # b = gr.Button("Submit")
-            b.click(fn=calc_mos, inputs=inputs, outputs=outputs, api_name="Submit")
+        # b = gr.Button("Submit")
+        b.click(fn=calc_mos, inputs=inputs, outputs=outputs, api_name="Submit")
 
         # Logger
         callback.setup(
@@ -488,10 +502,11 @@ with gr.Blocks(css=css, theme=my_theme) as demo:
                 hyp,
                 wer,
                 ppm,
-                msg],
+                msg,
+                saved_google_drive_id],
             flagging_dir="./exp/%s" % config["exp_id"],
         )
-
+        # Saving the Recording to CSV Logger (TO BE DELETED)
         with gr.Row():
             b2 = gr.Button("2. Save the Recording", variant="primary", elem_id="save")
             js_confirmed_saving = "(x) => confirm('Recording Saved!')"
@@ -508,6 +523,7 @@ with gr.Blocks(css=css, theme=my_theme) as demo:
                     wer,
                     ppm,
                     msg,
+                    saved_google_drive_id
                 ],
                 outputs=None,
                 preprocess=False,
@@ -527,9 +543,10 @@ with gr.Blocks(css=css, theme=my_theme) as demo:
                     wer,
                     ppm,
                     msg,
+                    saved_google_drive_id
                 ],
                 value="3.Clear All",
                 elem_id="clear",
             )
 
-demo.launch(share=True)
+demo.launch(share=False)

config/Arthur.yaml

@@ -3,10 +3,10 @@ ref_txt: data/Arthur_the_rat.txt
 ref_feature: data/Patient_sil_trim_16k_normed_5_snr_40/Arthur_the_rat.csv
 ref_wavs: data/Patient_sil_trim_16k_normed_5_snr_40/Arthur_the_rat
 thre:
-  minppm: 100
-  maxppm: 100
-  WER: 0.1
-  AUTOMOS: 4.0
+  minppm: 300
+  maxppm: 300
+  WER: 0.5
+  AUTOMOS: 2.0
 auth:
   username: Kath
   password: Kath

local/check_data.py

@@ -2,9 +2,11 @@ from googleapiclient.discovery import build
 from google.oauth2 import service_account
 from googleapiclient.http import MediaFileUpload
 import pdb
-pdb.set_trace()
 
 import gradio as gr
+'''
+Usage: python loacl/checkdata.py <google_file_id>
+'''
 
 # 来自Google Cloud控制台的JSON凭据文件
 credentials_file =  "./src/peerless-window-254907-b386b71c0d99.json"
@@ -22,13 +24,14 @@ files = results.get('files', [])
 print(files)
 from googleapiclient.http import MediaIoBaseDownload
 import io 
+import sys
 
-file_id = "1YjON2ObGM826KaaqF-sKM7CO0tAtzWGg"
+file_id = sys.argv[1]
+# "1YjON2ObGM826KaaqF-sKM7CO0tAtzWGg"
 # Get the file's metadata
 file = service.files().get(fileId=file_id).execute()
 
-pdb.set_trace()
-request = service.files().get_media(fileId="1YjON2ObGM826KaaqF-sKM7CO0tAtzWGg")
+request = service.files().get_media(fileId=file_id)
 with open(file['name'], 'wb') as file_obj:
     downloader = MediaIoBaseDownload(file_obj, request)
     done = False
@@ -38,6 +41,4 @@ with open(file['name'], 'wb') as file_obj:
 
 print(f"Downloaded: {file['name']}")
 
-pdb.set_trace()
-        
-# print('文件ID：%s' % response.get('id'))
+pdb.set_trace()

local/convert_metrics.py

@@ -0,0 +1,73 @@
+import numpy as np
+import matplotlib.pyplot as plt
+
+# Natural MOS to AVA MOS
+
+def linear_function(x):
+    m = (4 - 1) / (1.5 - 1)  
+    b = 1 - m * 1  
+    return m * x + b
+
+def quadratic_function(x):
+    return -0.0816 * (x - 5) ** 2 + 5
+
+# Natural MOS to AVA MOS
+def nat2avaMOS(x):
+    if x <= 1.5:
+        return linear_function(x)
+    elif x >1.5 and x <= 5:
+        return quadratic_function(x)
+
+# Word error rate to Intellibility Score (X is percentage)
+def WER2INTELI(x):
+    if x <= 10:
+        return 100
+    elif x <= 100:
+        slope = (30 - 100) / (100 - 10)
+        intercept = 100 - slope * 10
+        return slope * x + intercept
+    else:
+        return 100 * np.exp(-0.01 * (x - 100))
+
+# 生成 x 值
+# x = np.linspace(0, 200, 400)  # 从0到200生成400个点
+
+# 计算对应的 y 值
+# y = [WER2INTELI(xi) for xi in x]
+
+
+# plt.plot(x, y)
+# plt.xlabel('x')
+# plt.ylabel('f(x)')
+# plt.title('Custom Function')
+# plt.grid(True)
+# plt.show()
+
+# 生成 x 值的范围
+x1 = np.linspace(1, 1.5, 100)
+x2 = np.linspace(1.5, 5, 100)
+
+# 计算对应的 y 值
+y1 = linear_function(x1)
+y2 = quadratic_function(x2)
+
+# 绘制线性部分
+plt.plot(x1, y1, label='Linear Function (1 <= x <= 1.5)')
+
+# 绘制二次部分
+plt.plot(x2, y2, label='Quadratic Function (1.5 <= x <= 5)')
+
+# 添加标签和标题
+plt.xlabel('Natural Mean Opinion Score')
+plt.ylabel('AVA Mean Opinion Score')
+plt.title('nat2avaMOS')
+
+# 添加图例
+plt.legend()
+
+# 显示图形
+plt.grid(True)
+
+# 显示图像
+# plt.savefig("./local/nat2avaMOS.png")
+# plt.savefig("./local/WER2INT.png")