bugfixes and added auth

.gitignore

@@ -1,7 +1,7 @@
+.env
 *.json
 mapping
 *.ipynb
 test.py
-results/
 .notebook/
 __pycache__/

app.py

@@ -25,12 +25,13 @@ if not os.path.exists("results"):
     os.mkdir("results")
 
 if not os.path.exists(SAVE_PATH):
-    open(SAVE_PATH,"w").close()
+    with open(SAVE_PATH,"w") as f:
+        f.write("""email,path,Ori Apex_score,Ori Apex XT_score,deepgram_score,Ori Swift_score,Ori Prime_score,Ori Apex_appearance,Ori Apex XT_appearance,deepgram_appearance,Ori Swift_appearance,Ori Prime_appearance,Ori Apex_duration,Ori Apex XT_duration,deepgram_duration,Ori Swift_duration,Ori Prime_duration,azure_score,azure_appearance,azure_duration\n""")
 
 if not os.path.exists(TEMP_DIR):
     os.mkdir(TEMP_DIR)
 
-CREATE_TASK_URL = "https://ai-voice-test.voicegenie.ai/task"
+CREATE_TASK_URL = os.getenv("CREATE_TASK_URL")
 
 def decode_audio_array(base64_string):
     bytes_data = base64.b64decode(base64_string)
@@ -41,7 +42,10 @@ def decode_audio_array(base64_string):
     return audio_array
 
 def send_task(payload):
-    response = requests.post(CREATE_TASK_URL,json=payload)
+    header = {
+        "Authorization": f"Bearer {os.getenv('CREATE_TASK_API_KEY')}"
+    }
+    response = requests.post(CREATE_TASK_URL,json=payload,headers=header)
     response = response.json()
 
     if payload["task"] == "transcribe_with_fastapi":
@@ -591,76 +595,79 @@ def dashboard():
         st.title('Model Arena Scoreboard')
 
         df = pd.read_csv(SAVE_PATH)
-        metrics = calculate_metrics(df)
-
-        MODEL_DESCRIPTIONS = {
-            "Ori Prime": "Foundational, large, and stable.",
-            "Ori Swift": "Lighter and faster than Ori Prime.",
-            "Ori Apex": "The top-performing model, fast and stable.",
-            "Ori Apex XT": "Enhanced with more training, though slightly less stable than Ori Apex.",
-            "DG" : "Deepgram Nova-2 API",
-            "Azure" : "Azure Speech Services API"
-        }
+        if len(df) != 0:
+            metrics = calculate_metrics(df)
+
+            MODEL_DESCRIPTIONS = {
+                "Ori Prime": "Foundational, large, and stable.",
+                "Ori Swift": "Lighter and faster than Ori Prime.",
+                "Ori Apex": "The top-performing model, fast and stable.",
+                "Ori Apex XT": "Enhanced with more training, though slightly less stable than Ori Apex.",
+                "DG" : "Deepgram Nova-2 API",
+                "Azure" : "Azure Speech Services API"
+            }
 
-        st.header('Model Descriptions')
-
-        cols = st.columns(2)
-        for idx, (model, description) in enumerate(MODEL_DESCRIPTIONS.items()):
-            with cols[idx % 2]:
-                st.markdown(f"""
-                    <div style='padding: 1rem; border: 1px solid #e1e4e8; border-radius: 6px; margin-bottom: 1rem;'>
-                        <h3 style='margin: 0; margin-bottom: 0.5rem;'>{model}</h3>
-                        <p style='margin: 0; color: #6e7681;'>{description}</p>
-                    </div>
-                    """, unsafe_allow_html=True)
-
-        st.header('Overall Performance')
-
-        col1, col2, col3= st.columns(3)
-
-        with col1:
-            create_metric_container("Total Matches", len(df))
-
-        best_model = max(metrics.items(), key=lambda x: x[1]['win_rate'])[0]
-        with col2:
-            create_metric_container(
-                "Best Model",
-                get_model_abbreviation(best_model),
-                full_name=best_model
-            )
-
-        most_appearances = max(metrics.items(), key=lambda x: x[1]['appearances'])[0]
-        with col3:
-            create_metric_container(
-                "Most Used",
-                get_model_abbreviation(most_appearances),
-                full_name=most_appearances
-            )
-
-        st.header('Win Rates')
-        win_rate_chart = create_win_rate_chart(metrics)
-        st.plotly_chart(win_rate_chart, use_container_width=True)
-
-        st.header('Appearance Distribution')
-        appearance_chart = create_appearance_chart(metrics)
-        st.plotly_chart(appearance_chart, use_container_width=True)
-
-        st.header('Head-to-Head Analysis')
-        matrix_chart = create_head_to_head_matrix(df)
-        st.plotly_chart(matrix_chart, use_container_width=True)
-
-        st.header('Detailed Metrics')
-        metrics_df = pd.DataFrame.from_dict(metrics, orient='index')
-        metrics_df['win_rate'] = metrics_df['win_rate'].round(2)
-        metrics_df.drop(["avg_response_time","response_time_std"],axis=1,inplace=True)
-        # metrics_df['avg_response_time'] = metrics_df['avg_response_time'].round(3)
-        metrics_df.index = [get_model_abbreviation(model) for model in metrics_df.index]
-        st.dataframe(metrics_df)
-
-        st.header('Full Dataframe')
-        df = df.drop('path', axis=1)
-        df = df.drop(['Ori Apex_duration', 'Ori Apex XT_duration', 'deepgram_duration', 'Ori Swift_duration', 'Ori Prime_duration','azure_duration','email'],axis=1)
-        st.dataframe(df)
+            st.header('Model Descriptions')
+
+            cols = st.columns(2)
+            for idx, (model, description) in enumerate(MODEL_DESCRIPTIONS.items()):
+                with cols[idx % 2]:
+                    st.markdown(f"""
+                        <div style='padding: 1rem; border: 1px solid #e1e4e8; border-radius: 6px; margin-bottom: 1rem;'>
+                            <h3 style='margin: 0; margin-bottom: 0.5rem;'>{model}</h3>
+                            <p style='margin: 0; color: #6e7681;'>{description}</p>
+                        </div>
+                        """, unsafe_allow_html=True)
+
+            st.header('Overall Performance')
+
+            col1, col2, col3= st.columns(3)
+
+            with col1:
+                create_metric_container("Total Matches", len(df))
+
+            best_model = max(metrics.items(), key=lambda x: x[1]['win_rate'])[0]
+            with col2:
+                create_metric_container(
+                    "Best Model",
+                    get_model_abbreviation(best_model),
+                    full_name=best_model
+                )
+
+            most_appearances = max(metrics.items(), key=lambda x: x[1]['appearances'])[0]
+            with col3:
+                create_metric_container(
+                    "Most Used",
+                    get_model_abbreviation(most_appearances),
+                    full_name=most_appearances
+                )
+
+            st.header('Win Rates')
+            win_rate_chart = create_win_rate_chart(metrics)
+            st.plotly_chart(win_rate_chart, use_container_width=True)
+
+            st.header('Appearance Distribution')
+            appearance_chart = create_appearance_chart(metrics)
+            st.plotly_chart(appearance_chart, use_container_width=True)
+
+            st.header('Head-to-Head Analysis')
+            matrix_chart = create_head_to_head_matrix(df)
+            st.plotly_chart(matrix_chart, use_container_width=True)
+
+            st.header('Detailed Metrics')
+            metrics_df = pd.DataFrame.from_dict(metrics, orient='index')
+            metrics_df['win_rate'] = metrics_df['win_rate'].round(2)
+            metrics_df.drop(["avg_response_time","response_time_std"],axis=1,inplace=True)
+            # metrics_df['avg_response_time'] = metrics_df['avg_response_time'].round(3)
+            metrics_df.index = [get_model_abbreviation(model) for model in metrics_df.index]
+            st.dataframe(metrics_df)
+
+            st.header('Full Dataframe')
+            df = df.drop('path', axis=1)
+            df = df.drop(['Ori Apex_duration', 'Ori Apex XT_duration', 'deepgram_duration', 'Ori Swift_duration', 'Ori Prime_duration','azure_duration','email'],axis=1)
+            st.dataframe(df)
+        else:
+            st.write("No Data to show")
     else:
         st.write('You have not entered your email and name yet')
         st.write('Please Navigate to login page in the dropdown menu')
@@ -732,7 +739,7 @@ def validate_name(name):
     return re.match(pattern, name) is not None
 
 def create_login_page():
-    st.title("Welcome to the App")
+    st.title("Welcome to the Speech-To-Text Arena")
 
     if 'logged_in' not in st.session_state:
         st.session_state.logged_in = False

results/results.csv

@@ -0,0 +1 @@
+email,path,Ori Apex_score,Ori Apex XT_score,deepgram_score,Ori Swift_score,Ori Prime_score,Ori Apex_appearance,Ori Apex XT_appearance,deepgram_appearance,Ori Swift_appearance,Ori Prime_appearance,Ori Apex_duration,Ori Apex XT_duration,deepgram_duration,Ori Swift_dur