Testing tts

src/ui/Forms/Main.cs

@@ -19027,7 +19027,7 @@ namespace Nikse.SubtitleEdit.Forms
             }
             else if (e.Modifiers == (Keys.Alt | Keys.Shift | Keys.Control) && e.KeyCode == Keys.T)
             {
-                using (var form = new TextToSpeech(_subtitle, _videoFileName))
+                using (var form = new TextToSpeech(_subtitle, _videoFileName, _videoInfo))
                 {
                     if (form.ShowDialog(this) != DialogResult.OK)
                     {

src/ui/Forms/TextToSpeech.Designer.cs

@@ -29,9 +29,15 @@
         private void InitializeComponent()
         {
             this.buttonOK = new System.Windows.Forms.Button();
-            this.labelDescription = new System.Windows.Forms.Label();
+            this.labelProgress = new System.Windows.Forms.Label();
-            this.button1 = new System.Windows.Forms.Button();
+            this.buttonGenerateTTS = new System.Windows.Forms.Button();
-            this.textBox1 = new Nikse.SubtitleEdit.Controls.NikseTextBox();
+            this.progressBar1 = new System.Windows.Forms.ProgressBar();
+            this.labelEngine = new System.Windows.Forms.Label();
+            this.groupBoxMsSettings = new System.Windows.Forms.GroupBox();
+            this.labelMsVoice = new System.Windows.Forms.Label();
+            this.nikseComboBoxVoice = new Nikse.SubtitleEdit.Controls.NikseComboBox();
+            this.nikseComboBoxEngine = new Nikse.SubtitleEdit.Controls.NikseComboBox();
+            this.groupBoxMsSettings.SuspendLayout();
             this.SuspendLayout();
             // 
             // buttonOK
@@ -46,50 +52,144 @@
             this.buttonOK.UseVisualStyleBackColor = true;
             this.buttonOK.Click += new System.EventHandler(this.buttonOK_Click);
             // 
-            // labelDescription
+            // labelProgress
             // 
-            this.labelDescription.AutoSize = true;
+            this.labelProgress.AutoSize = true;
-            this.labelDescription.ImeMode = System.Windows.Forms.ImeMode.NoControl;
+            this.labelProgress.ImeMode = System.Windows.Forms.ImeMode.NoControl;
-            this.labelDescription.Location = new System.Drawing.Point(9, 33);
+            this.labelProgress.Location = new System.Drawing.Point(12, 384);
-            this.labelDescription.Name = "labelDescription";
+            this.labelProgress.Name = "labelProgress";
-            this.labelDescription.Size = new System.Drawing.Size(76, 13);
+            this.labelProgress.Size = new System.Drawing.Size(70, 13);
-            this.labelDescription.TabIndex = 9;
+            this.labelProgress.TabIndex = 9;
-            this.labelDescription.Text = "Name of voice";
+            this.labelProgress.Text = "labelProgress";
             // 
-            // button1
+            // buttonGenerateTTS
             // 
-            this.button1.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Bottom | System.Windows.Forms.AnchorStyles.Right)));
+            this.buttonGenerateTTS.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Bottom | System.Windows.Forms.AnchorStyles.Right)));
-            this.button1.ImeMode = System.Windows.Forms.ImeMode.NoControl;
+            this.buttonGenerateTTS.ImeMode = System.Windows.Forms.ImeMode.NoControl;
-            this.button1.Location = new System.Drawing.Point(12, 75);
+            this.buttonGenerateTTS.Location = new System.Drawing.Point(571, 374);
-            this.button1.Name = "button1";
+            this.buttonGenerateTTS.Name = "buttonGenerateTTS";
-            this.button1.Size = new System.Drawing.Size(75, 23);
+            this.buttonGenerateTTS.Size = new System.Drawing.Size(228, 23);
-            this.button1.TabIndex = 11;
+            this.buttonGenerateTTS.TabIndex = 11;
-            this.button1.Text = "Go";
+            this.buttonGenerateTTS.Text = "Generate speech from text";
-            this.button1.UseVisualStyleBackColor = true;
+            this.buttonGenerateTTS.UseVisualStyleBackColor = true;
-            this.button1.Click += new System.EventHandler(this.button1_Click);
+            this.buttonGenerateTTS.Click += new System.EventHandler(this.ButtonGenerateTtsClick);
             // 
-            // textBox1
+            // progressBar1
             // 
-            this.textBox1.Anchor = ((System.Windows.Forms.AnchorStyles)(((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Left) 
+            this.progressBar1.Anchor = ((System.Windows.Forms.AnchorStyles)(((System.Windows.Forms.AnchorStyles.Bottom | System.Windows.Forms.AnchorStyles.Left) 
             | System.Windows.Forms.AnchorStyles.Right)));
-            this.textBox1.FocusedColor = System.Drawing.Color.FromArgb(((int)(((byte)(0)))), ((int)(((byte)(120)))), ((int)(((byte)(215)))));
+            this.progressBar1.Location = new System.Drawing.Point(12, 407);
-            this.textBox1.Location = new System.Drawing.Point(12, 49);
+            this.progressBar1.Name = "progressBar1";
-            this.textBox1.Name = "textBox1";
+            this.progressBar1.Size = new System.Drawing.Size(706, 23);
-            this.textBox1.Size = new System.Drawing.Size(229, 20);
+            this.progressBar1.TabIndex = 12;
-            this.textBox1.TabIndex = 10;
+            // 
-            this.textBox1.Text = "Hello, how are you?";
+            // labelEngine
+            // 
+            this.labelEngine.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Right)));
+            this.labelEngine.AutoSize = true;
+            this.labelEngine.ImeMode = System.Windows.Forms.ImeMode.NoControl;
+            this.labelEngine.Location = new System.Drawing.Point(405, 12);
+            this.labelEngine.Name = "labelEngine";
+            this.labelEngine.Size = new System.Drawing.Size(40, 13);
+            this.labelEngine.TabIndex = 14;
+            this.labelEngine.Text = "Engine";
+            // 
+            // groupBoxMsSettings
+            // 
+            this.groupBoxMsSettings.Anchor = ((System.Windows.Forms.AnchorStyles)(((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Bottom) 
+            | System.Windows.Forms.AnchorStyles.Right)));
+            this.groupBoxMsSettings.Controls.Add(this.labelMsVoice);
+            this.groupBoxMsSettings.Controls.Add(this.nikseComboBoxVoice);
+            this.groupBoxMsSettings.Location = new System.Drawing.Point(408, 57);
+            this.groupBoxMsSettings.Name = "groupBoxMsSettings";
+            this.groupBoxMsSettings.Size = new System.Drawing.Size(391, 311);
+            this.groupBoxMsSettings.TabIndex = 15;
+            this.groupBoxMsSettings.TabStop = false;
+            this.groupBoxMsSettings.Text = "Settings";
+            // 
+            // labelMsVoice
+            // 
+            this.labelMsVoice.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Right)));
+            this.labelMsVoice.AutoSize = true;
+            this.labelMsVoice.ImeMode = System.Windows.Forms.ImeMode.NoControl;
+            this.labelMsVoice.Location = new System.Drawing.Point(14, 25);
+            this.labelMsVoice.Name = "labelMsVoice";
+            this.labelMsVoice.Size = new System.Drawing.Size(34, 13);
+            this.labelMsVoice.TabIndex = 16;
+            this.labelMsVoice.Text = "Voice";
+            // 
+            // nikseComboBoxVoice
+            // 
+            this.nikseComboBoxVoice.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Right)));
+            this.nikseComboBoxVoice.BackColor = System.Drawing.SystemColors.Window;
+            this.nikseComboBoxVoice.BackColorDisabled = System.Drawing.Color.FromArgb(((int)(((byte)(240)))), ((int)(((byte)(240)))), ((int)(((byte)(240)))));
+            this.nikseComboBoxVoice.BorderColor = System.Drawing.Color.FromArgb(((int)(((byte)(171)))), ((int)(((byte)(173)))), ((int)(((byte)(179)))));
+            this.nikseComboBoxVoice.BorderColorDisabled = System.Drawing.Color.FromArgb(((int)(((byte)(120)))), ((int)(((byte)(120)))), ((int)(((byte)(120)))));
+            this.nikseComboBoxVoice.ButtonForeColor = System.Drawing.SystemColors.ControlText;
+            this.nikseComboBoxVoice.ButtonForeColorDown = System.Drawing.Color.Orange;
+            this.nikseComboBoxVoice.ButtonForeColorOver = System.Drawing.Color.FromArgb(((int)(((byte)(0)))), ((int)(((byte)(120)))), ((int)(((byte)(215)))));
+            this.nikseComboBoxVoice.DropDownHeight = 400;
+            this.nikseComboBoxVoice.DropDownStyle = System.Windows.Forms.ComboBoxStyle.DropDownList;
+            this.nikseComboBoxVoice.DropDownWidth = 195;
+            this.nikseComboBoxVoice.FormattingEnabled = false;
+            this.nikseComboBoxVoice.Location = new System.Drawing.Point(17, 41);
+            this.nikseComboBoxVoice.MaxLength = 32767;
+            this.nikseComboBoxVoice.Name = "nikseComboBoxVoice";
+            this.nikseComboBoxVoice.SelectedIndex = -1;
+            this.nikseComboBoxVoice.SelectedItem = null;
+            this.nikseComboBoxVoice.SelectedText = "";
+            this.nikseComboBoxVoice.Size = new System.Drawing.Size(351, 23);
+            this.nikseComboBoxVoice.TabIndex = 15;
+            this.nikseComboBoxVoice.UsePopupWindow = false;
+            // 
+            // nikseComboBoxEngine
+            // 
+            this.nikseComboBoxEngine.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Right)));
+            this.nikseComboBoxEngine.BackColor = System.Drawing.SystemColors.Window;
+            this.nikseComboBoxEngine.BackColorDisabled = System.Drawing.Color.FromArgb(((int)(((byte)(240)))), ((int)(((byte)(240)))), ((int)(((byte)(240)))));
+            this.nikseComboBoxEngine.BorderColor = System.Drawing.Color.FromArgb(((int)(((byte)(171)))), ((int)(((byte)(173)))), ((int)(((byte)(179)))));
+            this.nikseComboBoxEngine.BorderColorDisabled = System.Drawing.Color.FromArgb(((int)(((byte)(120)))), ((int)(((byte)(120)))), ((int)(((byte)(120)))));
+            this.nikseComboBoxEngine.ButtonForeColor = System.Drawing.SystemColors.ControlText;
+            this.nikseComboBoxEngine.ButtonForeColorDown = System.Drawing.Color.Orange;
+            this.nikseComboBoxEngine.ButtonForeColorOver = System.Drawing.Color.FromArgb(((int)(((byte)(0)))), ((int)(((byte)(120)))), ((int)(((byte)(215)))));
+            this.nikseComboBoxEngine.DropDownHeight = 400;
+            this.nikseComboBoxEngine.DropDownStyle = System.Windows.Forms.ComboBoxStyle.DropDown;
+            this.nikseComboBoxEngine.DropDownWidth = 195;
+            this.nikseComboBoxEngine.FormattingEnabled = false;
+            this.nikseComboBoxEngine.Location = new System.Drawing.Point(408, 28);
+            this.nikseComboBoxEngine.MaxLength = 32767;
+            this.nikseComboBoxEngine.Name = "nikseComboBoxEngine";
+            this.nikseComboBoxEngine.SelectedIndex = -1;
+            this.nikseComboBoxEngine.SelectedItem = null;
+            this.nikseComboBoxEngine.SelectedText = "";
+            this.nikseComboBoxEngine.Size = new System.Drawing.Size(391, 23);
+            this.nikseComboBoxEngine.TabIndex = 13;
+            this.nikseComboBoxEngine.TabStop = false;
+            this.nikseComboBoxEngine.Text = "nikseComboBox1";
+            this.nikseComboBoxEngine.UsePopupWindow = false;
+            this.nikseComboBoxEngine.SelectedIndexChanged += new System.EventHandler(this.nikseComboBoxEngine_SelectedIndexChanged);
             // 
             // TextToSpeech
             // 
             this.AutoScaleDimensions = new System.Drawing.SizeF(6F, 13F);
             this.AutoScaleMode = System.Windows.Forms.AutoScaleMode.Font;
             this.ClientSize = new System.Drawing.Size(811, 442);
-            this.Controls.Add(this.button1);
+            this.Controls.Add(this.groupBoxMsSettings);
-            this.Controls.Add(this.textBox1);
+            this.Controls.Add(this.labelEngine);
-            this.Controls.Add(this.labelDescription);
+            this.Controls.Add(this.nikseComboBoxEngine);
+            this.Controls.Add(this.progressBar1);
+            this.Controls.Add(this.buttonGenerateTTS);
+            this.Controls.Add(this.labelProgress);
             this.Controls.Add(this.buttonOK);
+            this.KeyPreview = true;
+            this.MinimumSize = new System.Drawing.Size(827, 481);
             this.Name = "TextToSpeech";
-            this.Text = "TextToSpeach";
+            this.ShowIcon = false;
+            this.ShowInTaskbar = false;
+            this.StartPosition = System.Windows.Forms.FormStartPosition.CenterParent;
+            this.Text = "Text to speach";
+            this.groupBoxMsSettings.ResumeLayout(false);
+            this.groupBoxMsSettings.PerformLayout();
             this.ResumeLayout(false);
             this.PerformLayout();
 
@@ -97,8 +197,13 @@
 
         #endregion
         private System.Windows.Forms.Button buttonOK;
-        private Controls.NikseTextBox textBox1;
+        private System.Windows.Forms.Label labelProgress;
-        private System.Windows.Forms.Label labelDescription;
+        private System.Windows.Forms.Button buttonGenerateTTS;
-        private System.Windows.Forms.Button button1;
+        private System.Windows.Forms.ProgressBar progressBar1;
+        private Controls.NikseComboBox nikseComboBoxEngine;
+        private System.Windows.Forms.Label labelEngine;
+        private System.Windows.Forms.GroupBox groupBoxMsSettings;
+        private System.Windows.Forms.Label labelMsVoice;
+        private Controls.NikseComboBox nikseComboBoxVoice;
     }
 }

src/ui/Forms/TextToSpeech.cs

@@ -1,5 +1,10 @@
-using Nikse.SubtitleEdit.Logic;
+using Nikse.SubtitleEdit.Core.Common;
+using Nikse.SubtitleEdit.Logic;
 using System;
+using System.Collections.Generic;
+using System.Diagnostics;
+using System.IO;
+using System.Linq;
 using System.Speech.Synthesis;
 using System.Windows.Forms;
 
@@ -7,33 +12,322 @@ namespace Nikse.SubtitleEdit.Forms
 {
     public partial class TextToSpeech : Form
     {
-        public TextToSpeech(Core.Common.Subtitle subtitle, string videoFileName)
+        private readonly Subtitle _subtitle;
+        private readonly string _videoFileName;
+        private readonly VideoInfo _videoInfo;
+        private string _waveFolder;
+
+        public TextToSpeech(Subtitle subtitle, string videoFileName, VideoInfo videoInfo)
         {
             UiUtil.PreInitialize(this);
             InitializeComponent();
             UiUtil.FixFonts(this);
 
+            _subtitle = subtitle;
+            _videoFileName = videoFileName;
+            _videoInfo = videoInfo;
+
             buttonOK.Text = LanguageSettings.Current.General.Ok;
             UiUtil.FixLargeFonts(this, buttonOK);
+
+            progressBar1.Visible = false;
+            labelProgress.Text = string.Empty;
+
+            nikseComboBoxEngine.DropDownStyle = ComboBoxStyle.DropDownList;
+            nikseComboBoxEngine.Items.Clear();
+            nikseComboBoxEngine.Items.Add("Microsoft SpeechSynthesizer (fast/robotic)");
+            nikseComboBoxEngine.Items.Add("Tortoise TTS (very slow/very good)");
+            nikseComboBoxEngine.Items.Add("Mimic3");
+            nikseComboBoxEngine.SelectedIndex = 0;
         }
 
-        private void button1_Click(object sender, EventArgs e)
+        private void ButtonGenerateTtsClick(object sender, EventArgs e)
         {
-            var text = textBox1.Text;
+            _waveFolder = Path.Combine(Path.GetTempPath(), Guid.NewGuid().ToString());
+            Directory.CreateDirectory(_waveFolder);
 
-            using (SpeechSynthesizer synth = new SpeechSynthesizer())
+            if (nikseComboBoxEngine.SelectedIndex == 0)
             {
-                synth.SetOutputToWaveFile(@"C:\data\Sample.wav");
+                GenerateParagraphAudioMs();
-                synth.SelectVoiceByHints(VoiceGender.Male, VoiceAge.Adult);
-                var builder = new PromptBuilder();
-                builder.AppendText(text);
-                synth.Speak(builder);
             }
+            else if (nikseComboBoxEngine.SelectedIndex == 1)
+            {
+                GenerateParagraphAudioTortoiseTts();
+            }
+
+            var fileNames = FixParagraphAudioSpeed();
+
+            // rename result file
+            var tempAudioFile = MergeAudioParagraphs(fileNames);
+            var resultAudioFileName = Path.Combine(Path.GetDirectoryName(tempAudioFile), Path.GetFileNameWithoutExtension(_videoFileName) + ".wav");
+            File.Move(tempAudioFile, resultAudioFileName);
+            
+            Cleanup(_waveFolder, resultAudioFileName);
+
+            UiUtil.OpenFolder(_waveFolder);
+        }
+
+        private static void Cleanup(string waveFolder, string resultAudioFile)
+        {
+            foreach (var fileName in Directory.GetFiles(waveFolder, "*.wav"))
+            {
+                if (!fileName.Equals(resultAudioFile, StringComparison.OrdinalIgnoreCase))
+                {
+                    File.Delete(fileName);
+                }
+            }
+        }
+
+        private List<string> FixParagraphAudioSpeed()
+        {
+            var fileNames = new List<string>(_subtitle.Paragraphs.Count);
+
+            labelProgress.Text = "Adjusting speed...";
+            labelProgress.Refresh();
+            Application.DoEvents();
+
+            progressBar1.Value = 0;
+            progressBar1.Maximum = _subtitle.Paragraphs.Count;
+            progressBar1.Visible = true;
+            for (var index = 0; index < _subtitle.Paragraphs.Count; index++)
+            {
+                progressBar1.Value = index + 1;
+                labelProgress.Text = $"Adjusting speed: {index + 1} / {_subtitle.Paragraphs.Count}...";
+                var p = _subtitle.Paragraphs[index];
+                var next = _subtitle.GetParagraphOrDefault(index + 1);
+                var pFileName = Path.Combine(_waveFolder, index + ".wav");
+
+                var addDuration = 0d;
+                if (next != null && p.EndTime.TotalMilliseconds < next.StartTime.TotalMilliseconds)
+                {
+                    var diff = next.StartTime.TotalMilliseconds - p.EndTime.TotalMilliseconds;
+                    addDuration = Math.Max(1000, diff);
+                }
+
+                var outputFileName1 = Path.Combine(_waveFolder, index + "_u.wav");
+                var trimProcess = VideoPreviewGenerator.TrimSilenceStartAndEnd(pFileName, outputFileName1);
+                trimProcess.Start();
+                trimProcess.WaitForExit();
+
+                var waveInfo = UiUtil.GetVideoInfo(outputFileName1);
+                if (waveInfo.TotalMilliseconds <= p.DurationTotalMilliseconds + addDuration)
+                {
+                    fileNames.Add(outputFileName1);
+                    continue;
+                }
+
+                var factor = waveInfo.TotalMilliseconds / (p.DurationTotalMilliseconds + addDuration);
+                var outputFileName2 = Path.Combine(_waveFolder, index + "_t.wav");
+                fileNames.Add(outputFileName2);
+                var mergeProcess = VideoPreviewGenerator.ChangeSpeed(outputFileName1, outputFileName2, (float)factor);
+                mergeProcess.Start();
+                mergeProcess.WaitForExit();
+            }
+
+            return fileNames;
+        }
+
+        private string MergeAudioParagraphs(List<string> fileNames)
+        {
+            labelProgress.Text = "Merging audio track...";
+            labelProgress.Refresh();
+            Application.DoEvents();
+            var silenceFileName = Path.Combine(_waveFolder, "silence.wav");
+            var silenceProcess = VideoPreviewGenerator.GenerateEmptyAudio(silenceFileName, (float)_videoInfo.TotalSeconds);
+            silenceProcess.Start();
+            silenceProcess.WaitForExit();
+
+            progressBar1.Value = 0;
+            progressBar1.Maximum = _subtitle.Paragraphs.Count;
+            progressBar1.Visible = true;
+            var inputFileName = silenceFileName;
+            var outputFileName = string.Empty;
+            for (var index = 0; index < fileNames.Count; index++)
+            {
+                progressBar1.Value = index + 1;
+                labelProgress.Text = $"Merging audio track: {index + 1} / {_subtitle.Paragraphs.Count}...";
+                var p = _subtitle.Paragraphs[index];
+                var pFileName = fileNames[index];
+                outputFileName = Path.Combine(_waveFolder, $"silence{index}.wav");
+                var mergeProcess = VideoPreviewGenerator.MergeAudioTracks(inputFileName, pFileName, outputFileName, (float)p.StartTime.TotalSeconds);
+                inputFileName = outputFileName;
+                mergeProcess.Start();
+                mergeProcess.WaitForExit();
+            }
+
+            progressBar1.Visible = false;
+            labelProgress.Text = string.Empty;
+
+            return outputFileName;
+        }
+
+        private void GenerateParagraphAudioMs()
+        {
+            progressBar1.Value = 0;
+            progressBar1.Maximum = _subtitle.Paragraphs.Count;
+            progressBar1.Visible = true;
+
+            using (var synthesizer = new SpeechSynthesizer())
+            {
+                VoiceInfo voiceInfo = null;
+                var vs = synthesizer.GetInstalledVoices().Where(p => p.Enabled).ToList();
+                for (var index = 0; index < vs.Count; index++)
+                {
+                    var v = vs[index];
+                    if (index == nikseComboBoxVoice.SelectedIndex)
+                    {
+                        synthesizer.SelectVoice(v.VoiceInfo.Name);
+                        voiceInfo = v.VoiceInfo;
+                    }
+                }
+
+                for (var index = 0; index < _subtitle.Paragraphs.Count; index++)
+                {
+                    progressBar1.Value = index + 1;
+                    labelProgress.Text = $"Generating audio texts: {index + 1} / {_subtitle.Paragraphs.Count}...";
+                    var p = _subtitle.Paragraphs[index];
+                    synthesizer.SetOutputToWaveFile(Path.Combine(_waveFolder, index + ".wav"));
+                    var builder = new PromptBuilder();
+                    if (voiceInfo != null)
+                    {
+                        builder.StartVoice(voiceInfo);
+                    }
+
+                    builder.AppendText(p.Text);
+                    if (voiceInfo != null)
+                    {
+                        builder.EndVoice();
+                    }
+
+                    synthesizer.Speak(builder);
+
+                    progressBar1.Refresh();
+                    labelProgress.Refresh();
+                    Application.DoEvents();
+                }
+            }
+
+            progressBar1.Visible = false;
+            labelProgress.Text = string.Empty;
+        }
+
+        private bool GenerateParagraphAudioTortoiseTts()
+        {
+            var pythonFolder = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.UserProfile),
+                "AppData",
+                "Local",
+                "Programs",
+                "Python");
+
+            var pyFileName = "do_tts.py";
+            var voice = nikseComboBoxVoice.Text;
+            var files = Directory.EnumerateFiles(pythonFolder, pyFileName, SearchOption.AllDirectories).ToList();
+            if (files.Count == 0)
+            {
+                MessageBox.Show($"{pyFileName} not found under {pythonFolder}");
+                return false;
+            }
+
+            var pythonFolderVersionFolder = files[0].Substring(pythonFolder.Length).Split(new char[] { Path.DirectorySeparatorChar }, StringSplitOptions.RemoveEmptyEntries);
+            var pythonExe = Path.Combine(pythonFolder, pythonFolderVersionFolder[0], "python.exe");
+            if (!File.Exists(pythonExe))
+            {
+                pythonExe = "python.exe";
+            }
+
+            progressBar1.Value = 0;
+            progressBar1.Maximum = _subtitle.Paragraphs.Count;
+            progressBar1.Visible = true;
+
+
+            for (var index = 0; index < _subtitle.Paragraphs.Count; index++)
+            {
+                progressBar1.Value = index + 1;
+                labelProgress.Text = $"Generating audio texts: {index + 1} / {_subtitle.Paragraphs.Count}...";
+                var p = _subtitle.Paragraphs[index];
+                var outputFileName = Path.Combine(_waveFolder, index + ".wav");
+
+                var processTortoiseTts = new Process
+                {
+                    StartInfo =
+                    {
+                        WorkingDirectory = Path.GetDirectoryName(files[0]),
+                        FileName = pythonExe,
+                        Arguments = $"do_tts.py --output_path \"{_waveFolder}\" --preset ultra_fast --voice {voice} --text \"{p.Text.RemoveChar('"')}\"",
+                        UseShellExecute = false,
+                        CreateNoWindow = true,
+                    }
+                };
+
+                processTortoiseTts.Start();
+                processTortoiseTts.WaitForExit();
+
+                var inputFile = Path.Combine(_waveFolder, $"{voice}_0_2.wav");
+                File.Move(inputFile, outputFileName);
+
+                progressBar1.Refresh();
+                labelProgress.Refresh();
+                Application.DoEvents();
+            }
+
+            progressBar1.Visible = false;
+            labelProgress.Text = string.Empty;
+
+            return true;
         }
 
         private void buttonOK_Click(object sender, EventArgs e)
         {
             DialogResult = DialogResult.OK;
         }
+
+        private void nikseComboBoxEngine_SelectedIndexChanged(object sender, EventArgs e)
+        {
+            nikseComboBoxVoice.Items.Clear();
+
+            if (nikseComboBoxEngine.SelectedIndex == 0)
+            {
+                using (var synthesizer = new SpeechSynthesizer())
+                {
+                    foreach (var v in synthesizer.GetInstalledVoices())
+                    {
+                        if (v.Enabled)
+                        {
+                            nikseComboBoxVoice.Items.Add(v.VoiceInfo.Name);
+                        }
+                    }
+                }
+            }
+
+            if (nikseComboBoxEngine.SelectedIndex == 1)
+            {
+                nikseComboBoxVoice.Items.Add("angie");
+                nikseComboBoxVoice.Items.Add("applejack");
+                nikseComboBoxVoice.Items.Add("daniel");
+                nikseComboBoxVoice.Items.Add("deniro");
+                nikseComboBoxVoice.Items.Add("emma");
+                nikseComboBoxVoice.Items.Add("freeman");
+                nikseComboBoxVoice.Items.Add("geralt");
+                nikseComboBoxVoice.Items.Add("halle");
+                nikseComboBoxVoice.Items.Add("jlaw");
+                nikseComboBoxVoice.Items.Add("lj");
+                nikseComboBoxVoice.Items.Add("mol");
+                nikseComboBoxVoice.Items.Add("myself");
+                nikseComboBoxVoice.Items.Add("pat");
+                nikseComboBoxVoice.Items.Add("pat2");
+                nikseComboBoxVoice.Items.Add("rainbow");
+                nikseComboBoxVoice.Items.Add("Update rainbow");
+                nikseComboBoxVoice.Items.Add("snakes");
+                nikseComboBoxVoice.Items.Add("tim_reynolds");
+                nikseComboBoxVoice.Items.Add("tom");
+                nikseComboBoxVoice.Items.Add("weaver");
+                nikseComboBoxVoice.Items.Add("william");
+            }
+
+            if (nikseComboBoxVoice.Items.Count > 0)
+            {
+                nikseComboBoxVoice.SelectedIndex = 0;
+            }
+        }
     }
 }

src/ui/Logic/VideoPreviewGenerator.cs

@@ -75,6 +75,42 @@ namespace Nikse.SubtitleEdit.Logic
             return processMakeVideo;
         }
 
+        public static Process GenerateEmptyAudio(string outputFileName, float seconds, DataReceivedEventHandler dataReceivedHandler = null)
+        {
+            var processMakeVideo = new Process
+            {
+                StartInfo =
+                {
+                    FileName = GetFfmpegLocation(),
+                    Arguments = $"-f lavfi -i anullsrc -t {seconds.ToString(CultureInfo.InvariantCulture)} \"{outputFileName}\"",
+                    UseShellExecute = false,
+                    CreateNoWindow = true
+                }
+            };
+
+            SetupDataReceiveHandler(dataReceivedHandler, processMakeVideo);
+
+            return processMakeVideo;
+        }
+
+        public static Process MergeAudioTracks(string inputFileName1, string inputFileName2, string outputFileName, float startSeconds, DataReceivedEventHandler dataReceivedHandler = null)
+        {
+            var processMakeVideo = new Process
+            {
+                StartInfo =
+                {
+                    FileName = GetFfmpegLocation(),
+                    Arguments = $"-i \"{inputFileName1}\" -i \"{inputFileName2}\" -filter_complex \"aevalsrc=0:d={startSeconds.ToString(CultureInfo.InvariantCulture)}[s1];[s1][1:a]concat=n=2:v=0:a=1[ac1];[0:a][ac1]amix=2:normalize=false[aout]\" -map [aout] \"{outputFileName}\"",
+                    UseShellExecute = false,
+                    CreateNoWindow = true
+                }
+            };
+
+            SetupDataReceiveHandler(dataReceivedHandler, processMakeVideo);
+
+            return processMakeVideo;
+        }
+
         private static void SetupDataReceiveHandler(DataReceivedEventHandler dataReceivedHandler, Process processMakeVideo)
         {
             if (dataReceivedHandler != null)
@@ -546,5 +582,45 @@ namespace Nikse.SubtitleEdit.Logic
             SetupDataReceiveHandler(outputHandler, processMakeVideo);
             return processMakeVideo;
         }
+
+        public static Process ChangeSpeed(string inputFileName, string outputFileName, float inputSpeed, DataReceivedEventHandler dataReceivedHandler = null)
+        {
+            var speed = Math.Max(0.5f, inputSpeed);
+            speed = Math.Min(100, speed);
+            speed = (float)Math.Round(speed, 3, MidpointRounding.AwayFromZero);
+
+            var processMakeVideo = new Process
+            {
+                StartInfo =
+                {
+                    FileName = GetFfmpegLocation(),
+                    Arguments = $"-i \"{inputFileName}\" -filter:a \"atempo={speed.ToString(CultureInfo.InvariantCulture)}\" \"{outputFileName}\"",
+                    UseShellExecute = false,
+                    CreateNoWindow = true
+                }
+            };
+
+            SetupDataReceiveHandler(dataReceivedHandler, processMakeVideo);
+
+            return processMakeVideo;
+        }
+
+        public static Process TrimSilenceStartAndEnd(string inputFileName, string outputFileName, DataReceivedEventHandler dataReceivedHandler = null)
+        {
+            var processMakeVideo = new Process
+            {
+                StartInfo =
+                {
+                    FileName = GetFfmpegLocation(),
+                    Arguments = $"-i \"{inputFileName}\" -af silenceremove=start_periods=1:start_silence=0.01,areverse,silenceremove=start_periods=1:start_silence=0.01,areverse \"{outputFileName}\"",
+                    UseShellExecute = false,
+                    CreateNoWindow = true
+                }
+            };
+
+            SetupDataReceiveHandler(dataReceivedHandler, processMakeVideo);
+
+            return processMakeVideo;
+        }
     }
 }