Update Purfview's Whisper-Faster to r145.3

2024-11-22 19:22:53 +01:00 · 2023-08-18 19:47:39 +02:00 · 2023-08-18 19:47:39 +02:00 · d4c735b80e
commit d4c735b80e
parent b7ed8c98c6
4 changed files with 44 additions and 33 deletions
--- a/src/ui/Forms/AudioToText/WhisperAdvanced.Designer.cs
+++ b/src/ui/Forms/AudioToText/WhisperAdvanced.Designer.cs
@ -43,7 +43,7 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
            this.tabPageOpenAI = new System.Windows.Forms.TabPage();
            this.textBoxOpenAI = new System.Windows.Forms.TextBox();
            this.tabPageFasterWhisper = new System.Windows.Forms.TabPage();
-            this.textBox1 = new System.Windows.Forms.TextBox();
+            this.textBoxPurfviewFasterWhisper = new System.Windows.Forms.TextBox();
            this.comboBoxWhisperExtra = new Nikse.SubtitleEdit.Controls.NikseComboBox();
            this.tabControlCommandLineHelp.SuspendLayout();
            this.TabPageCPP.SuspendLayout();
@ -188,7 +188,7 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
            // 
            // tabPageFasterWhisper
            // 
-            this.tabPageFasterWhisper.Controls.Add(this.textBox1);
+            this.tabPageFasterWhisper.Controls.Add(this.textBoxPurfviewFasterWhisper);
            this.tabPageFasterWhisper.Location = new System.Drawing.Point(4, 22);
            this.tabPageFasterWhisper.Name = "tabPageFasterWhisper";
            this.tabPageFasterWhisper.Padding = new System.Windows.Forms.Padding(3);
@ -197,19 +197,19 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
            this.tabPageFasterWhisper.Text = "Faster Whisper";
            this.tabPageFasterWhisper.UseVisualStyleBackColor = true;
            // 
-            // textBox1
+            // textBoxPurfviewFasterWhisper
            // 
-            this.textBox1.Dock = System.Windows.Forms.DockStyle.Fill;
-            this.textBox1.Font = new System.Drawing.Font("Consolas", 8.25F, System.Drawing.FontStyle.Regular, System.Drawing.GraphicsUnit.Point, ((byte)(0)));
-            this.textBox1.Location = new System.Drawing.Point(3, 3);
-            this.textBox1.Multiline = true;
-            this.textBox1.Name = "textBox1";
-            this.textBox1.ReadOnly = true;
-            this.textBox1.ScrollBars = System.Windows.Forms.ScrollBars.Vertical;
-            this.textBox1.Size = new System.Drawing.Size(824, 413);
-            this.textBox1.TabIndex = 3;
-            this.textBox1.Text = resources.GetString("textBox1.Text");
-            this.textBox1.WordWrap = false;
+            this.textBoxPurfviewFasterWhisper.Dock = System.Windows.Forms.DockStyle.Fill;
+            this.textBoxPurfviewFasterWhisper.Font = new System.Drawing.Font("Consolas", 8.25F, System.Drawing.FontStyle.Regular, System.Drawing.GraphicsUnit.Point, ((byte)(0)));
+            this.textBoxPurfviewFasterWhisper.Location = new System.Drawing.Point(3, 3);
+            this.textBoxPurfviewFasterWhisper.Multiline = true;
+            this.textBoxPurfviewFasterWhisper.Name = "textBoxPurfviewFasterWhisper";
+            this.textBoxPurfviewFasterWhisper.ReadOnly = true;
+            this.textBoxPurfviewFasterWhisper.ScrollBars = System.Windows.Forms.ScrollBars.Vertical;
+            this.textBoxPurfviewFasterWhisper.Size = new System.Drawing.Size(824, 413);
+            this.textBoxPurfviewFasterWhisper.TabIndex = 3;
+            this.textBoxPurfviewFasterWhisper.Text = resources.GetString("textBoxPurfviewFasterWhisper.Text");
+            this.textBoxPurfviewFasterWhisper.WordWrap = false;
            // 
            // comboBoxWhisperExtra
            // 
@ -284,6 +284,6 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
        private System.Windows.Forms.TextBox textBoxConstMe;
        private System.Windows.Forms.TextBox textBoxOpenAI;
        private System.Windows.Forms.TabPage tabPageFasterWhisper;
-        private System.Windows.Forms.TextBox textBox1;
+        private System.Windows.Forms.TextBox textBoxPurfviewFasterWhisper;
    }
 }
--- a/src/ui/Forms/AudioToText/WhisperAdvanced.resx
+++ b/src/ui/Forms/AudioToText/WhisperAdvanced.resx
@ -229,24 +229,26 @@
                        (default: 0)
 </value>
  </data>
-  <data name="textBox1.Text" xml:space="preserve">
+  <data name="textBoxPurfviewFasterWhisper.Text" xml:space="preserve">
    <value>  --temperature TEMPERATURE
                        temperature to use for sampling (default: 0)
-  --best_of BEST_OF     number of candidates when sampling with non-zero temperature (default: 5)
-  --beam_size BEAM_SIZE
+  --best_of BEST_OF, -bo BEST_OF
+                        number of candidates when sampling with non-zero temperature (default: 5)
+  --beam_size BEAM_SIZE, -bs BEAM_SIZE
                        number of beams in beam search, only applicable when temperature is zero (default: 1)
-  --patience PATIENCE   optional patience value to use in beam decoding, as in https://arxiv.org/abs/2204.05424, the
+  --patience PATIENCE, -p PATIENCE
+                        optional patience value to use in beam decoding, as in https://arxiv.org/abs/2204.05424, the
                        default (1.0) is equivalent to conventional beam search (default: 1.0)
  --length_penalty LENGTH_PENALTY
                        optional token length penalty coefficient (alpha) as in https://arxiv.org/abs/1609.08144, uses
                        simple length normalization by default (default: 1.0)
  --repetition_penalty REPETITION_PENALTY
-                        Penalty applied to the score of previously generated tokens (set > 1.0 to
-                        penalize). (default: 1.0)
+                        Penalty applied to the score of previously generated tokens (set &gt; 1.0 to penalize). (default:
+                        1.0)
  --suppress_tokens SUPPRESS_TOKENS
                        comma-separated list of token ids to suppress during sampling; '-1' will suppress most special
                        characters except common punctuations (default: -1)
-  --initial_prompt INITIAL_PROMPT
+  --initial_prompt INITIAL_PROMPT, -prompt INITIAL_PROMPT
                        optional text to provide as a prompt for the first window. (default: None)
  --condition_on_previous_text CONDITION_ON_PREVIOUS_TEXT
                        if True, provide the previous output of the model as a prompt for the next window; disabling
@ -257,7 +259,7 @@
                        condition_on_previous_text is True. (default: 0.5)
  --temperature_increment_on_fallback TEMPERATURE_INCREMENT_ON_FALLBACK
                        temperature to increase when falling back when the decoding fails to meet either of the
-                        thresholds below (default: 0.2)
+                        thresholds below. To disable fallback set it to 'None'. (default: 0.2)
  --compression_ratio_threshold COMPRESSION_RATIO_THRESHOLD
                        if the gzip compression ratio is higher than this value, treat the decoding as failed
                        (default: 2.4)
@ -286,7 +288,7 @@
  --threads THREADS     number of threads used for CPU inference; By default number of the real cores but no more that
                        4 (default: 0)
  --version             Show Faster-Whisper's version number
-  --vad_filter VAD_FILTER
+  --vad_filter VAD_FILTER, -vad VAD_FILTER
                        Enable the voice activity detection (VAD) to filter out parts of the audio without speech.
                        (default: True)
  --vad_threshold VAD_THRESHOLD
@ -303,11 +305,18 @@
  --vad_window_size_samples VAD_WINDOW_SIZE_SAMPLES
                        Size of audio chunks fed to the silero VAD model. Values other than 512, 1024, 1536 may affect
                        model perfomance!!! (default: 1536)
-  --compute_type {default,auto,int8,int8_float16,int16,float16,float32,int8_bfloat16,bfloat16}
-                        Type of quantization to use (see https://opennmt.net/CTranslate2/quantization.html) (default:
+  --compute_type {default,auto,int8,int8_float16,int8_float32,int8_bfloat16,int16,float16,float32,bfloat16}, -ct {default,auto,int8,int8_float16,int8_float32,int8_bfloat16,int16,float16,float32,bfloat16}
+                        Type of quantization to use (see https://opennmt.net/CTranslate2/quantization.html). (default:
                        default)
-  --beep BEEP           Enables beep sound when operation is finished. (default: True)
-  --skip SKIP           Skips files if 'srt' subtitle exists. Works if input is wildcard or directory. (default:
-                        False)</value>
+  --batch_recursive, -br
+                        Enables recursive batch processing. Note: If set then it changes defaults of --output_dir.
+                        (default: False)
+  --beep_off            Disables beep sound when operation is finished. (default: False)
+  --skip                Skips files if 'srt' subtitle exists. Works if input is wildcard or directory. (default:
+                        False)
+  --checkcuda, -cc      Check for CUDA devices. (for Subtitle Edit's internal use)
+  --print_progress, -pp
+                        Prints progress bar instead of transcription. (default: False)
+	</value>
  </data>
-</root>
+</root>
--- a/src/ui/Forms/AudioToText/WhisperDownload.cs
+++ b/src/ui/Forms/AudioToText/WhisperDownload.cs
@ -60,15 +60,16 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
        };


-        private const string DownloadUrlPurfviewFasterWhisper = "https://github.com/Purfview/whisper-standalone-win/releases/download/faster-whisper/Whisper-Faster_r145.2.zip";
+        private const string DownloadUrlPurfviewFasterWhisper = "https://github.com/Purfview/whisper-standalone-win/releases/download/faster-whisper/Whisper-Faster_r145.3.zip";

        private static readonly string[] Sha512HashesPurfviewFasterWhisper =
        {
-            "b689f5ff7329f0ae8f08e3d42b1a2f71bcbe2717cf1231395791cf3b90e305ba4e92955a62ebe946a73c5ca83f61bc60b2e4cff1065cc0f49cfc1f3c665fa587", // r145.2 
+            "fee96c9f8f3a9b67c2e1923fa0f5ef388d645aa3788b1b00c9f12392ef2db4b905d84f5c00ab743a284c8ba2750121e08e9fee1edc76d9c0f12ae51d61b1b12a", // r145.3.zip
        };

        private static readonly string[] OldSha512HashesPurfviewFasterWhisper =
        {
+            "b689f5ff7329f0ae8f08e3d42b1a2f71bcbe2717cf1231395791cf3b90e305ba4e92955a62ebe946a73c5ca83f61bc60b2e4cff1065cc0f49cfc1f3c665fa587", // r145.2 
            "75ba2bcee9fef0846e54ce367df3fb54f3b9f4cb0f8ac33f01bdde44dc313cd01b3263b43c899271af5901f765ef6257358dcf68c11024652299942405afe289", //  r145.1
            "5414c15bb1682efc2f737f3ab5f15c4350a70c30a6101b631297420bbc4cb077ef9b88cb6e5512f4adcdafbda85eb894ff92eae07bd70c66efa0b28a08361033", // Whisper-Faster r141.4
        };
@ -312,7 +313,7 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText

            if (whisperChoice == WhisperChoice.PurfviewFasterWhisper)
            {
-                return hash == "ab34a7cb4ce36a0634756da9a2592d81281404c3408ca223dcd9f2bdbc131e81b8a45158d9df295b42fc2bce5485ae1542c00a76efe33775ef811fbef42e86e3";
+                return hash == "f55f3d7f60ec3a5ba6e1cc59062ad34d5e235f726f12bc8e4d1c5c519d1950fecd3e0386f87d937cabb7041235ef3cff54752d82de9f901690cb227ac4d2c2da";
            }

            return hash == "c43fed38d1ae99e6fbbd8c842c2d550b4949081c0c7fba72cd2e2e8435ff05eac4f64e659efb09d597c3c062edf1e5026acc375d2a07290fa3c0fca9ac3bd7a2";
--- a/src/ui/SubtitleEdit.csproj
+++ b/src/ui/SubtitleEdit.csproj
@ -1657,6 +1657,7 @@
    </EmbeddedResource>
    <EmbeddedResource Include="Forms\AudioToText\WhisperAdvanced.resx">
      <DependentUpon>WhisperAdvanced.cs</DependentUpon>
+      <SubType>Designer</SubType>
    </EmbeddedResource>
    <EmbeddedResource Include="Forms\AudioToText\WhisperDownload.resx">
      <DependentUpon>WhisperDownload.cs</DependentUpon>