First working version

annotations.csv

@@ -1,15 +1,4 @@
 user_id,gender,audio_file,score
-,[],audio1.wav,3.0
-,[],audio2.wav,3.0
 dkl.,Female,audio1.wav,3.0
 dkl.,Female,audio2.wav,3.0
-,Other,audio1.wav,1.0
-jhgs,Other,audio1.wav,3.0
-ermjb,Other,audio1.wav,
-ermjb,Other,audio2.wav,1.0
-gvsdg,Other,audio1.wav,
-gvsdg,Other,audio2.wav,
-dsfsd,Other,audio1.wav,
-dsfsd,Other,audio2.wav,1.0
-fdsfsd,Other,audio1.wav,3.0
 dfsd,Other,audio1.wav,2

app.py

@@ -2,6 +2,8 @@ import gradio as gr
 import pandas as pd
 import os
 import hashlib
+from datasets import Dataset, concatenate_datasets, load_dataset
+from huggingface_hub import login
 
 # ============================
 # CONFIGURATION
@@ -10,22 +12,44 @@ import hashlib
 AUDIO_FOLDER = "audio_files"
 audio_files = sorted(os.listdir(AUDIO_FOLDER))
 
-ANNOTATION_FILE = "annotations.csv"
-
 ADMIN_PASSWORD_HASH = "eb32da077dfaf326cd6f73e0716b628da6427aa318a2d0b9fafa9ef315b5e885"
 
 # ============================
-# STORAGE INITIALIZATION
+# HF DATASET CONFIGURATION
 # ============================
 
-if not os.path.exists(ANNOTATION_FILE):
-    df = pd.DataFrame(columns=["user_id", "gender", "audio_file", "score"])
-    df.to_csv(ANNOTATION_FILE, index=False)
+HF_DATASET_NAME = "MeysamSh/Annotate_Samples"  # replace with your HF dataset
+HF_TOKEN = os.environ.get("HF_TOKEN")  # store your token as a secret in Spaces
+login(HF_TOKEN)
 
 # ============================
 # USER FUNCTIONS
 # ============================
 
+def load_hf_dataset():
+    """Load existing HF Dataset or create empty one if not exists."""
+    try:
+        ds = load_dataset(HF_DATASET_NAME, split="train")
+    except:
+        # Dataset does not exist yet
+        df = pd.DataFrame(columns=["user_id", "gender", "audio_file", "score"])
+        ds = Dataset.from_pandas(df)
+        ds.push_to_hub(HF_DATASET_NAME, private=True)
+    return ds
+
+def submit_annotation_hf(user_id, gender, score, audio_file):
+    """Append annotation to HF Dataset."""
+    ds = load_hf_dataset()
+    new_row = pd.DataFrame([{
+        "user_id": user_id,
+        "gender": gender,
+        "audio_file": audio_file,
+        "score": score
+    }])
+    ds_new = Dataset.from_pandas(new_row)
+    ds_combined = concatenate_datasets([ds, ds_new])
+    ds_combined.push_to_hub(HF_DATASET_NAME)
+
 def load_next_audio(state):
     if state is None:
         state = {"index": 0}
@@ -33,13 +57,11 @@ def load_next_audio(state):
     idx = state["index"]
 
     if idx >= len(audio_files):
-        return None, state, False, "All audio files annotated."
+        return None, state, {"played": 0}, "All audio files annotated."
 
     filepath = os.path.join(AUDIO_FOLDER, audio_files[idx])
-    # return filepath, state, False, f"Loaded {audio_files[idx]}"
     return filepath, state, {"played": 0}, f"Loaded {audio_files[idx]}"
 
-
 def submit_annotation(user_id, gender, score, state):
     if state is None:
         state = {"index": 0}
@@ -51,14 +73,7 @@ def submit_annotation(user_id, gender, score, state):
 
     audio_file = audio_files[idx]
 
-    df = pd.read_csv(ANNOTATION_FILE)
-    df = pd.concat([df, pd.DataFrame([{
-        "user_id": user_id,
-        "gender": gender,
-        "audio_file": audio_file,
-        "score": score
-    }])], ignore_index=True)
-    df.to_csv(ANNOTATION_FILE, index=False)
+    submit_annotation_hf(user_id, gender, score, audio_file)
 
     state["index"] += 1
     return state, f"Saved rating for {audio_file}."
@@ -67,8 +82,8 @@ def submit_annotation(user_id, gender, score, state):
 # SUBMIT BUTTON CONTROL
 # ============================
 
-def check_submit_ready(user_id, audio_played,score):
-    ready = len(user_id.strip())>1 and audio_played['played'] == 1 and score != "None"
+def check_submit_ready(user_id, audio_played, score):
+    ready = len(user_id.strip()) > 1 and audio_played['played'] == 1 and score != "None"
     return gr.update(interactive=ready)
 
 def mark_audio_played():
@@ -83,11 +98,15 @@ def hash_password(password: str) -> str:
 
 def admin_login(input_password):
     if hash_password(input_password) == ADMIN_PASSWORD_HASH:
-        df = pd.read_csv(ANNOTATION_FILE)
+        ds = load_hf_dataset()
+        df = ds.to_pandas()
+        # For download: write temp CSV
+        temp_csv = "annotations_export.csv"
+        df.to_csv(temp_csv, index=False)
         return (
             gr.update(visible=True),
             df,
-            ANNOTATION_FILE,
+            temp_csv,
             gr.update(value="Admin authentication successful.")
         )
     else:
@@ -117,8 +136,7 @@ with gr.Blocks() as demo:
         user_id = gr.Textbox(label="User ID")
         gender = gr.Dropdown(["Male", "Female", "Other"], label="Gender", value="Other")
 
-    audio_player = gr.Audio(
-        label="Audio File")
+    audio_player = gr.Audio(label="Audio File")
 
     with gr.Row():
         score = gr.Dropdown(choices=["None","1", "2", "3", "4", "5"], value="None", label="MOS Score (1–5)")
@@ -143,19 +161,17 @@ with gr.Blocks() as demo:
     # Enable submit button only when conditions are met
     user_id.change(
         check_submit_ready,
-        inputs=[user_id, audio_played,score],
+        inputs=[user_id, audio_played, score],
         outputs=submit_btn
     )
-
     audio_played.change(
         check_submit_ready,
-        inputs=[user_id, audio_played,score],
+        inputs=[user_id, audio_played, score],
         outputs=submit_btn
     )
-
     score.change(
         check_submit_ready,
-        inputs=[user_id, audio_played,score],
+        inputs=[user_id, audio_played, score],
         outputs=submit_btn
     )
 
@@ -180,18 +196,15 @@ with gr.Blocks() as demo:
     gr.Markdown("## Admin Dashboard (Restricted Access)")
 
     with gr.Row():
-        admin_password = gr.Textbox(
-            label="Admin Password",
-            type="password"
-        )
+        admin_password = gr.Textbox(label="Admin Password", type="password")
         admin_login_btn = gr.Button("Login")
 
-    login_status = gr.Textbox(label="Login Status:",interactive=False)
+    login_status = gr.Textbox(label="Login Status:", interactive=False)
 
     with gr.Column(visible=False) as admin_panel:
         gr.Markdown("### Annotation Results")
-        results_table = gr.DataFrame(interactive=False,value=pd.read_csv(ANNOTATION_FILE))
-        download_admin = gr.File(label="Download annotations.csv",value=ANNOTATION_FILE)
+        results_table = gr.DataFrame(interactive=False, value=pd.DataFrame(columns=["user_id","gender","audio_file","score"]))
+        download_admin = gr.File(label="Download annotations.csv", value=None)
 
     admin_login_btn.click(
         admin_login,
@@ -199,7 +212,9 @@ with gr.Blocks() as demo:
         outputs=[admin_panel, results_table, download_admin, login_status]
     )
 
+# ============================
+# APP LAUNCH
+# ============================
+
 if __name__ == "__main__":
-    demo.launch(share=True)
-    # demo.launch(server_name="0.0.0.0", server_port=7860)
-    # demo.launch(server_name="0.0.0.0", server_port=7860, prevent_thread_lock=True)
+    demo.launch()