app.py

import gradio as gr
import pandas as pd
import librosa
import numpy as np
from datasets import Dataset, DatasetDict
class DataEditor:
    def __init__(self,df):
        self.df=df
        self.current_selected = -1
        self.current_page = 0
        self.datatable =df
        self.data =self.df[['text','flag']]
        self.sdata =self.df['audio'].to_list()
    def settt(self,df):
        self.df=pd.DataFrame()
        self.data =pd.DataFrame()
        self.sdata =[]
        self.df=df
        self.next_prveo=0
        self.data =self.df[['text','flag']]
        self.sdata =self.df['audio'].to_list()
        self.current_page = 0
        self.current_selected =1
        return self.data
    def get_prev_page(self,pagenumber):
        if self.next_prveo>=0:
            self.next_prveo-=1
            self.current_page=self.next_prveo
            row = self.data.iloc[self.next_prveo]
            txt_audio = row['text']
            return txt_audio
    def finsh_data(self):
        self.df['audio'] = self.sdata
        self.df[['text','flag']]=self.data
        return self.df
    def get_next_page(self,pagenumber):

          if self.next_prveo<9:
            self.next_prveo+=1
            self.current_page=self.next_prveo
            row = self.data.iloc[self.next_prveo]
            txt_audio = row['text']
            return txt_audio


    def get_page_data(self, page_number):
        start_index = page_number * 10
        end_index = start_index + 10
        return self.data.iloc[start_index:end_index]
    def update_page(self, new_page):
        self.current_page = new_page
        return (
            self.get_page_data(self.current_page),
            self.current_page > 0,
            self.current_page < len(self.data) // 10 - 1,
            self.current_page
        )
    def convert_dataframe_to_dataset(self, namedata):
        datatable=self.finsh_data()
        if "__index_level_0__" in datatable.columns:
               datatable =datatable.drop(columns=["__index_level_0__"])
        train_df =datatable
        ds = {
            "train": Dataset.from_pandas(train_df)
        }
        dataset = DatasetDict(ds)
        dirr = '/content/drive/MyDrive/vitsM/DATA/sata/NewData2hba/' + namedata
        dataset.save_to_disk(dirr)
        return dirr
    def read_dataset(self, link):
        try:
            dataset = DatasetDict.load_from_disk(link,token=os.environ.get("auth_acess_data"))
            df= dataset["train"].to_pandas()
            v=self.settt(df)
            return self.get_page_data(self.current_page),link

        except FileNotFoundError:
            return None, f"Error: Dataset not found at {link}"
        except Exception as e:
            return None, f"Error loading dataset: {e}"
    def on_select(self,evt:gr.SelectData):
        if evt.index:
            index_now = evt.index[0]
            self.current_selected = (self.current_page * 10) + index_now
            row = self.data.iloc[self.current_selected]
            txt_audio = row['text']
            #row_audio = self.sdata[self.current_selected]
            #if row_audio == None:
            zeros_array = np.zeros(100)
            audio_array = np.zeros(10, dtype=np.float32)
            audio = librosa.resample(audio_array, orig_sr=16000, target_sr=16000)
            #audio=librosa.resample(zeros_array,orig_sr=16000,target_sr=16000)
            return txt_audio,None
          #  else:
            return ""

        else:
            return None," "
    def on_saveAs_row(self):
        return self.get_page_data(self.current_page),None,""

    def on_row_save(self, text,data_oudio):
          if text!="" :

                row = self.data.iloc[self.current_selected]
                #row['text'] = text
                row['flag']=1
                self.data.iloc[self.current_selected] = row
                if data_oudio!=None:
                  self.sdata[self.current_selected] = data_oudio


                return  "",data_oudio
          else:
               return "",None
    def on_row_delete(self):
        if self.current_selected>=0:
              row = self.data.iloc[self.current_selected]
                #row['text'] = text
              row['flag']=0
              self.data.iloc[self.current_selected] = row
              self.sdata[self.current_selected] = None



        return self.get_page_data(self.current_page),None,""

    def create_interface(self):
        with gr.Blocks() as demo:
            with gr.Column():
                self.txturll = gr.Textbox(placeholder="link dir", interactive=True)
                self.btn_displayy = gr.Button("Load Dataset")
            with gr.Row():
                with gr.Column():
                    self.table = gr.Dataframe(value=self.datatable, headers=['text', 'audio'], interactive=True)
                    with gr.Row(equal_height=False):
                            self.prev_button = gr.Button("Previous Page",scale=1, size="sm")
                            self.page_number = gr.Number(value=self.current_page + 1, label="Page",scale=1)
                            self.next_button = gr.Button("Next Page",scale=1, size="sm")
                    with gr.Column():
                        self.txtsaveurl = gr.Textbox(placeholder="Save Dataset", interactive=True)
                        self.btn_savedataset = gr.Button("Save Dataset")
                        self.label =gr.Text("STATE")

                with gr.Column():

                     self.txt_audio = gr.Textbox(label="Audio Text", interactive=True,rtl=True)
                     self.btn_record = gr.Audio(interactive=True)

                     with gr.Row():
                          self.btn_save = gr.Button("Save")
                          self.btn_saveAs = gr.Button("SaveAs")
                         # self.buttonnext=gr.Button("Next")
                         # self.buttonprev=gr.Button("Prev")
                          self.btn_delete = gr.Button("Delete")


            self.btn_displayy.click(self.read_dataset, [self.txturll], [self.table, self.txtsaveurl])
            self.table.select(self.on_select, None,[self.txt_audio,self.btn_record])
            self.btn_save.click(self.on_row_save, [self.txt_audio,self.btn_record],[self.txt_audio,self.btn_record])
            self.btn_saveAs.click(self.on_saveAs_row, [], [self.table, self.txt_audio, self.label])
            self.btn_delete.click(self.on_row_delete, [], [self.table, self.txt_audio, self.label])
            #self.buttonnext.click(lambda page:self.get_next_page(page+1), [self.page_number], [self.txt_audio])
            #self.buttonprev.click(lambda page:self.get_prev_page(page-1), [self.page_number], [self.txt_audio])
            self.btn_savedataset.click(self.convert_dataframe_to_dataset, [self.txtsaveurl], [self.label])
            self.prev_button.click(lambda page: self.update_page(page - 1), [self.page_number], [self.table, self.prev_button, self.next_button, self.page_number])
                   #self.btn_save.click(self.save_row, [self.txt_audio,self.audio_player], [self.data_table])
            self.next_button.click(lambda page: self.update_page(page + 1), [self.page_number], [self.table, self.prev_button, self.next_button, self.page_number])
        demo.launch()
df1=pd.DataFrame(columns=['text','flag','audio'])
editor = DataEditor(df1)
editor.create_interface()