add audio player

components/dashboard_page.py

@@ -1,146 +1,88 @@
-import os
-from pathlib import Path
-
 import gradio as gr
 import numpy as np
-from pydub import AudioSegment
-
 from components.header import Header
 from utils.logger import Logger
+from utils.gdrive_downloader import PublicFolderAudioLoader
+from config import conf
 
 log = Logger()
-
-# اگر فایل‌های صوتی در پوشهٔ خاصی هستند این را عوض کنید
-AUDIO_DIR = Path("audio")  #  <project_root>/audio/<filename>.wav
+LOADER = PublicFolderAudioLoader(conf.GDRIVE_API_KEY)
+GDRIVE_FOLDER = conf.GDRIVE_FOLDER
 
 
 class DashboardPage:
-    """صفحهٔ داشبورد شامل اطلاعات متنی (چپ) و پخش‌کنندهٔ صوت (راست)."""
-
-    # ───────── ساخت UI ───────── #
     def __init__(self) -> None:
         with gr.Column(visible=False) as self.container:
-            # هدر
             self.header = Header()
 
-            # بدنهٔ دو ستونه
             with gr.Row():
-
-                # -------- ستونهٔ چپ : متادیتا -------- #
-                with gr.Column(scale=3) as self.left_col:
-
+                # ستون چپ
+                with gr.Column(scale=3):
                     with gr.Row():
                         self.tts_id = gr.Textbox(label="ID", interactive=False)
                         self.filename = gr.Textbox(label="Filename", interactive=False)
-                        
                     self.sentence = gr.Textbox(
                         label="Sentence", interactive=False, max_lines=5, rtl=True
                     )
-                    
                     self.ann_sentence = gr.Textbox(
                         label="Annotated Sentence",
                         interactive=True,
                         max_lines=5,
                         rtl=True,
                     )
-
                     with gr.Row():
                         self.ann_at = gr.Textbox(
-                            label="Annotation Time",
-                            interactive=False,
+                            label="Annotation Time", interactive=False
                         )
-
                         self.validated = gr.Checkbox(
-                            label="Annotation is Validate",
-                            interactive=False,
+                            label="Validated", interactive=False
                         )
-
-                    # دکمه‌های پیمایش زیر اطلاعات متنی
                     with gr.Row():
                         self.btn_prev = gr.Button("⬅️ Previous")
                         self.btn_next = gr.Button("Next ➡️")
 
-                # -------- ستونهٔ راست : پخش‌کننده -------- #
-                with gr.Column(scale=2) as self.right_col:
+                # ستون راست
+                with gr.Column(scale=2):
+                    self.btn_load_voice = gr.Button("Load Audio")
                     self.audio = gr.Audio(label="🔊 Audio", interactive=False)
-                    
 
-            # stateهای مخفی
-            self.items_state = gr.State([])  # list[dict]
-            self.idx_state = gr.State(0)  # اندیس فعلی
+            # stateها
+            self.items_state = gr.State([])
+            self.idx_state = gr.State(0)
 
-    # ───────── wiring ───────── #
+    # ---------------- wiring ---------------- #
     def register_callbacks(
-        self,
-        login_page,
-        session_state: gr.State,  # dict درون gr.State
-        root_blocks: gr.Blocks,
-    ) -> None:
+        self, login_page, session_state: gr.State, root_blocks: gr.Blocks
+    ):
 
-        # رویداد خروج
         self.header.register_callbacks(login_page, self, session_state)
 
-        # ---------- helpers ---------- #
-        def _audio_path(filename: str) -> str:
-            """مسیر کامل فایل صوتی روی دیسک."""
-            return str(AUDIO_DIR / filename)
-
-        def _duration_seconds(wav_path: str) -> float:
-            """طول فایل صوتی به ثانیه (برای اسلایدرها)."""
-            try:
-                dur = len(AudioSegment.from_file(wav_path)) / 1000.0
-                return round(dur, 2)
-            except Exception as e:
-                log.warning(f"Cannot read duration for '{wav_path}': {e}")
-                return 0.0
-
-        def show_current(items: list, idx: int):
-            """داده‌های رکورد idx را برای خروجی‌ها تولید می‌کند."""
+        # ---- helpers ----
+        def show_current(items, idx):
             if not items:
-                # 6 فیلد متنی + 3 فیلد صوت + validated
-                return [
-                    "",
-                    "",
-                    "",
-                    "",
-                    "",
-                    False,
-                    None,
-                    gr.update(minimum=0, maximum=0, value=0),
-                    gr.update(minimum=0, maximum=0, value=0),
-                ]
-
-            data = items[idx]
-            wav_path = _audio_path(data["filename"])
-            dur = _duration_seconds(wav_path)
-
+                return ["", "", "", "", "", False, None]
+            d = items[idx]
             return [
-                data["id"],
-                data["filename"],
-                data["sentence"],
-                data.get("annotated_sentence", ""),
-                data.get("annotated_at", ""),
-                bool(data.get("validated", False)),
-                wav_path,  # audio
-                gr.update(minimum=0, maximum=dur, value=0),  # start slider
-                gr.update(minimum=0, maximum=dur, value=dur),  # end slider
+                d["id"],
+                d["filename"],
+                d["sentence"],
+                d.get("annotated_sentence", ""),
+                d.get("annotated_at", ""),
+                bool(d.get("validated", False)),
+                None,  # Audio هنوز لود نشده
             ]
 
-        def next_idx(items: list, idx: int):
+        def next_idx(items, idx):
             return min(idx + 1, max(len(items) - 1, 0))
 
-        def prev_idx(items: list, idx: int):
+        def prev_idx(items, idx):
             return max(idx - 1, 0)
 
-        # ---------- initial load ---------- #
-        def load_items(sess: dict):
+        def load_items(sess):
             items = sess.get("dashboard_items", [])
-            return (
-                items,
-                0,
-                *show_current(items, 0),
-            )
+            return items, 0, *show_current(items, 0)
 
+        # ---------- initial load ---------- #
         root_blocks.load(
             fn=load_items,
             inputs=[session_state],
@@ -153,17 +95,15 @@ class DashboardPage:
                 self.ann_sentence,
                 self.ann_at,
                 self.validated,
-                self.audio,
-                self.start_slider,
-                self.end_slider,
+                self.audio,  # 9 اُمین خروجى
             ],
         )
 
-        # ---------- prev / next buttons ---------- #
-        for btn, fn_nav in [(self.btn_prev, prev_idx), (self.btn_next, next_idx)]:
+        # ---------- navigation ---------- #
+        for btn, nav_fn in [(self.btn_prev, prev_idx), (self.btn_next, next_idx)]:
             (
                 btn.click(
-                    fn=fn_nav,
+                    fn=nav_fn,
                     inputs=[self.items_state, self.idx_state],
                     outputs=self.idx_state,
                 ).then(
@@ -177,36 +117,49 @@ class DashboardPage:
                         self.ann_at,
                         self.validated,
                         self.audio,
-                        self.start_slider,
-                        self.end_slider,
                     ],
                 )
             )
 
-        # ---------- Play-Selection button ---------- #
-        def play_selection(wav_path: str, start: float, end: float):
-            """
-            بخش انتخاب‌شده از فایل را جدا می‌کند و به‌صورت
-            (sr, np.array) برمی‌گرداند تا در Player پخش شود.
-            """
-            if not wav_path or not os.path.exists(wav_path):
+        # ---------- load audio ---------- #
+        def disable_buttons():
+            return [
+                gr.update(value="⏳ Waiting...", interactive=False),
+                gr.update(interactive=False),
+                gr.update(interactive=False),
+            ]
+
+        def enable_buttons():
+            return [
+                gr.update(value="Load Audio", interactive=True),
+                gr.update(interactive=True),
+                gr.update(interactive=True),
+            ]
+
+        def download_voice(folder_link, filename):
+            if not filename:
                 return None
             try:
-                seg = AudioSegment.from_file(wav_path)
-                start_ms = int(max(start, 0) * 1000)
-                end_ms = int(min(end, len(seg) / 1000) * 1000)
-                if start_ms >= end_ms:
-                    end_ms = start_ms + 1000  # حداقل ۱ ثانیه
-                clip = seg[start_ms:end_ms]
-                samples = np.array(clip.get_array_of_samples()).astype(np.float32)
-                samples /= np.iinfo(samples.dtype).max  # نرمال‌سازی
-                return (clip.frame_rate, samples)
+                sr, wav = LOADER.load_audio(folder_link, filename)
+                return (sr, wav)
             except Exception as e:
-                log.error(f"Cannot slice audio '{wav_path}': {e}")
-                return None
-
-        self.play_btn.click(
-            fn=play_selection,
-            inputs=[self.audio, self.start_slider, self.end_slider],
-            outputs=self.audio,
+                log.error(f"GDrive download failed: {e}")
+                return None  # player خالي مى‌ماند
+
+        (
+            self.btn_load_voice.click(
+                fn=disable_buttons,
+                inputs=None,
+                outputs=[self.btn_load_voice, self.btn_prev, self.btn_next],
+            )
+            .then(
+                fn=download_voice,
+                inputs=[gr.State(GDRIVE_FOLDER), self.filename],
+                outputs=self.audio,
+            )
+            .then(
+                fn=enable_buttons,
+                inputs=None,
+                outputs=[self.btn_load_voice, self.btn_prev, self.btn_next],
+            )
         )

config.py

@@ -12,7 +12,8 @@ class Config(BaseSettings):
     DB_NAME: str = os.getenv("DB_NAME", "defaultdb")
     HF_TOKEN: str = os.environ.get("HF_TOKEN")
     HF_TTS_DS_REPO: str = os.environ.get("HF_TTS_DS_REPO")
-    GOOGLE_DRIVE_API_KEY: str = os.environ.get("GOOGLE_DRIVE_API_KEY") 
+    GDRIVE_API_KEY: str = os.environ.get("GDRIVE_API_KEY") 
+    GDRIVE_FOLDER: str = os.environ.get("GDRIVE_FOLDER") 
 
     APP_TITLE: str = "Gooya TTS Annotation Tools"
 

scripts/assign_interval_to_annotator.py

@@ -7,9 +7,9 @@ from utils.logger import Logger
 
 log = Logger()
 
-START_IDX = 1
-END_IDX = 100
-ANNOTATOR_NAME = "navid"
+START_IDX = 101
+END_IDX = 120
+ANNOTATOR_NAME = "vargha"
 
 with get_db() as db:
     annot_repo = AnnotatorRepo(db)

test/db_test.py

@@ -1,26 +0,0 @@
-import mysql.connector
-from mysql.connector import Error
-import config 
-
-try:
-    conn = mysql.connector.connect(**config.db_config)
-    if conn.is_connected():
-        print("✅OK!")
-
-        cursor = conn.cursor()
-        cursor.execute("SELECT id, filename, sentence FROM tts_data ORDER BY id DESC LIMIT 5")
-        records = cursor.fetchall()
-
-        print("🟢Last 5 records:")
-        for record in records:
-            print(record)
-
-except Error as e:
-    print("❌ error:", e)
-
-finally:
-    if 'cursor' in locals():
-        cursor.close()
-    if 'conn' in locals() and conn.is_connected():
-        conn.close()
-        print("⛔️Closed.")

gdrive_test.py → test/gdrive_test.py

@@ -1,8 +1,8 @@
 import gradio as gr
-from test.gdrive_downloader import PublicFolderAudioLoader
+from utils.gdrive_downloader import PublicFolderAudioLoader
 from config import conf
 
-LOADER  = PublicFolderAudioLoader(conf.GOOGLE_DRIVE_API_KEY)
+LOADER  = PublicFolderAudioLoader(conf.GDRIVE_API_KEY)
 
 def fetch_audio(folder_link, filename):
     sr, wav = LOADER.load_audio(folder_link, filename)
@@ -12,7 +12,7 @@ demo = gr.Interface(
     fn=fetch_audio,
     inputs=[
         gr.Textbox(label="Folder URL or ID",
-                   value="https://drive.google.com/drive/folders/15UllyqvOB8zmhzsTL8f1wmnK4OY2nzUQ?usp=sharing"),
+                   value=conf.GDRIVE_FOLDER),
         gr.Textbox(label="Filename (e.g. 0001.wav)")
     ],
     outputs=gr.Audio(label="🔊 Audio"),