Mirrors/SubtitleEdit
1
0
Fork 0
mirror of https://github.com/SubtitleEdit/subtitleedit.git synced 2024-11-21 18:52:36 +01:00
Code Issues Projects Releases Wiki Activity

Work on #8069

Browse Source
This commit is contained in:
Nikolaj Olsson 2024-03-25 07:51:41 +01:00
parent 19eb96a2b4
commit 7dd36ae09f
3 changed files with 160 additions and 82 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

62
src/ui/Forms/AudioToText/WhisperAdvanced.Designer.cs generated
View File

@ -43,8 +43,12 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
this.tabPageOpenAI = new System.Windows.Forms.TabPage();
this.textBoxOpenAI = new Nikse.SubtitleEdit.Controls.NikseTextBox();
this.tabPageFasterWhisper = new System.Windows.Forms.TabPage();
this.buttonStandard = new System.Windows.Forms.Button();
this.buttonSentence = new System.Windows.Forms.Button();
this.buttonSingleWords = new System.Windows.Forms.Button();
this.textBoxPurfviewFasterWhisper = new Nikse.SubtitleEdit.Controls.NikseTextBox();
this.comboBoxWhisperExtra = new Nikse.SubtitleEdit.Controls.NikseComboBox();
this.button1 = new System.Windows.Forms.Button();
this.tabControlCommandLineHelp.SuspendLayout();
this.TabPageCPP.SuspendLayout();
this.tabPageConstMe.SuspendLayout();
@ -191,6 +195,10 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
//
// tabPageFasterWhisper
//
this.tabPageFasterWhisper.Controls.Add(this.button1);
this.tabPageFasterWhisper.Controls.Add(this.buttonStandard);
this.tabPageFasterWhisper.Controls.Add(this.buttonSentence);
this.tabPageFasterWhisper.Controls.Add(this.buttonSingleWords);
this.tabPageFasterWhisper.Controls.Add(this.textBoxPurfviewFasterWhisper);
this.tabPageFasterWhisper.Location = new System.Drawing.Point(4, 22);
this.tabPageFasterWhisper.Name = "tabPageFasterWhisper";
@ -200,9 +208,44 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
this.tabPageFasterWhisper.Text = "Faster Whisper";
this.tabPageFasterWhisper.UseVisualStyleBackColor = true;
//
// buttonStandard
//
this.buttonStandard.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Bottom | System.Windows.Forms.AnchorStyles.Left)));
this.buttonStandard.Location = new System.Drawing.Point(311, 390);
this.buttonStandard.Name = "buttonStandard";
this.buttonStandard.Size = new System.Drawing.Size(148, 23);
this.buttonStandard.TabIndex = 6;
this.buttonStandard.Text = "Standard";
this.buttonStandard.UseVisualStyleBackColor = true;
this.buttonStandard.Click += new System.EventHandler(this.buttonStandard_Click);
//
// buttonSentence
//
this.buttonSentence.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Bottom | System.Windows.Forms.AnchorStyles.Left)));
this.buttonSentence.Location = new System.Drawing.Point(157, 390);
this.buttonSentence.Name = "buttonSentence";
this.buttonSentence.Size = new System.Drawing.Size(148, 23);
this.buttonSentence.TabIndex = 5;
this.buttonSentence.Text = "Sentence";
this.buttonSentence.UseVisualStyleBackColor = true;
this.buttonSentence.Click += new System.EventHandler(this.buttonSentence_Click);
//
// buttonSingleWords
//
this.buttonSingleWords.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Bottom | System.Windows.Forms.AnchorStyles.Left)));
this.buttonSingleWords.Location = new System.Drawing.Point(3, 390);
this.buttonSingleWords.Name = "buttonSingleWords";
this.buttonSingleWords.Size = new System.Drawing.Size(148, 23);
this.buttonSingleWords.TabIndex = 4;
this.buttonSingleWords.Text = "Single word";
this.buttonSingleWords.UseVisualStyleBackColor = true;
this.buttonSingleWords.Click += new System.EventHandler(this.buttonSingleWords_Click);
//
// textBoxPurfviewFasterWhisper
//
this.textBoxPurfviewFasterWhisper.Dock = System.Windows.Forms.DockStyle.Fill;
this.textBoxPurfviewFasterWhisper.Anchor = ((System.Windows.Forms.AnchorStyles)((((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Bottom)
| System.Windows.Forms.AnchorStyles.Left)
| System.Windows.Forms.AnchorStyles.Right)));
this.textBoxPurfviewFasterWhisper.FocusedColor = System.Drawing.Color.FromArgb(((int)(((byte)(0)))), ((int)(((byte)(120)))), ((int)(((byte)(215)))));
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);
@ -210,7 +253,7 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
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.Size = new System.Drawing.Size(824, 381);
this.textBoxPurfviewFasterWhisper.TabIndex = 3;
this.textBoxPurfviewFasterWhisper.Text = resources.GetString("textBoxPurfviewFasterWhisper.Text");
this.textBoxPurfviewFasterWhisper.WordWrap = false;
@ -240,6 +283,17 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
this.comboBoxWhisperExtra.UsePopupWindow = false;
this.comboBoxWhisperExtra.KeyDown += new System.Windows.Forms.KeyEventHandler(this.comboBoxWhisperExtra_KeyDown);
//
// button1
//
this.button1.Anchor = ((System.Windows.Forms.AnchorStyles)((System.Windows.Forms.AnchorStyles.Bottom | System.Windows.Forms.AnchorStyles.Left)));
this.button1.Location = new System.Drawing.Point(465, 390);
this.button1.Name = "button1";
this.button1.Size = new System.Drawing.Size(180, 23);
this.button1.TabIndex = 7;
this.button1.Text = "Highlight current word";
this.button1.UseVisualStyleBackColor = true;
this.button1.Click += new System.EventHandler(this.button1_Click);
//
// WhisperAdvanced
//
this.AutoScaleDimensions = new System.Drawing.SizeF(6F, 13F);
@ -290,5 +344,9 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
private Nikse.SubtitleEdit.Controls.NikseTextBox textBoxOpenAI;
private System.Windows.Forms.TabPage tabPageFasterWhisper;
private Nikse.SubtitleEdit.Controls.NikseTextBox textBoxPurfviewFasterWhisper;
private Button buttonSingleWords;
private Button buttonStandard;
private Button buttonSentence;
private Button button1;
}
}

20
src/ui/Forms/AudioToText/WhisperAdvanced.cs
View File

@ -115,5 +115,25 @@ namespace Nikse.SubtitleEdit.Forms.AudioToText
comboBoxWhisperExtra.Focus();
comboBoxWhisperExtra.SelectAll();
}
private void buttonSingleWords_Click(object sender, EventArgs e)
{
comboBoxWhisperExtra.Text = "--one_word 1";
}
private void buttonSentence_Click(object sender, EventArgs e)
{
comboBoxWhisperExtra.Text = "--sentence";
}
private void buttonStandard_Click(object sender, EventArgs e)
{
comboBoxWhisperExtra.Text = "--standard";
}
private void button1_Click(object sender, EventArgs e)
{
comboBoxWhisperExtra.Text = "--highlight_words true";
}
}
}

160
src/ui/Forms/AudioToText/WhisperAdvanced.resx
View File

@ -117,6 +117,86 @@
<resheader name="writer">
<value>System.Resources.ResXResourceWriter, System.Windows.Forms, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089</value>
</resheader>
<data name="textBoxPurfviewFasterWhisper.Text" xml:space="preserve">
<value>[--device DEVICE]
[--output_dir OUTPUT_DIR]
[--output_format {lrc,txt,text,vtt,srt,tsv,json,all}]
[--verbose VERBOSE]
[--task {transcribe,translate}]
[--temperature TEMPERATURE]
[--best_of BEST_OF]
[--beam_size BEAM_SIZE]
[--patience PATIENCE]
[--length_penalty LENGTH_PENALTY]
[--repetition_penalty REPETITION_PENALTY]
[--no_repeat_ngram_size NO_REPEAT_NGRAM_SIZE]
[--suppress_blank SUPPRESS_BLANK]
[--suppress_tokens SUPPRESS_TOKENS]
[--initial_prompt INITIAL_PROMPT]
[--prefix PREFIX]
[--condition_on_previous_text CONDITION_ON_PREVIOUS_TEXT]
[--prompt_reset_on_temperature PROMPT_RESET_ON_TEMPERATURE]
[--without_timestamps WITHOUT_TIMESTAMPS]
[--max_initial_timestamp MAX_INITIAL_TIMESTAMP]
[--temperature_increment_on_fallback TEMPERATURE_INCREMENT_ON_FALLBACK]
[--compression_ratio_threshold COMPRESSION_RATIO_THRESHOLD]
[--logprob_threshold LOGPROB_THRESHOLD]
[--no_speech_threshold NO_SPEECH_THRESHOLD]
[--hallucination_silence_threshold HALLUCINATION_SILENCE_THRESHOLD]
[--clip_timestamps CLIP_TIMESTAMPS]
[--no_speech_strict_lvl {0,1,2}]
[--word_timestamps WORD_TIMESTAMPS]
[--highlight_words HIGHLIGHT_WORDS]
[--prepend_punctuations PREPEND_PUNCTUATIONS]
[--append_punctuations APPEND_PUNCTUATIONS]
[--threads THREADS]
[--version]
[--vad_filter VAD_FILTER]
[--vad_threshold VAD_THRESHOLD]
[--vad_min_speech_duration_ms VAD_MIN_SPEECH_DURATION_MS]
[--vad_max_speech_duration_s VAD_MAX_SPEECH_DURATION_S]
[--vad_min_silence_duration_ms VAD_MIN_SILENCE_DURATION_MS]
[--vad_speech_pad_ms VAD_SPEECH_PAD_MS]
[--vad_window_size_samples VAD_WINDOW_SIZE_SAMPLES]
[--max_new_tokens MAX_NEW_TOKENS]
[--chunk_length CHUNK_LENGTH]
[--compute_type {default,auto,int8,int8_float16,int8_float32,int8_bfloat16,int16,float16,float32,bfloat16}]
[--batch_recursive]
[--beep_off]
[--skip]
[--checkcuda]
[--print_progress]
[--postfix]
[--check_files]
[--PR163_off]
[--hallucinations_list_off]
[--one_word {0,1,2}]
[--sentence]
[--standard]
[--standard_asia]
[--max_comma MAX_COMMA]
[--max_comma_cent {50,60,70,80,90,100}]
[--max_gap MAX_GAP]
[--max_line_width MAX_LINE_WIDTH]
[--max_line_count MAX_LINE_COUNT]
[--min_dist_to_end {0,4,5,6,7,8,9,10,11,12}]
[--prompt_max {16,32,64,128,223}]
[--reprompt {0,1,2}]
[--prompt_reset_on_no_end {0,1,2}]
[--ff_dump]
[--ff_mp3]
[--ff_sync]
[--ff_rnndn_sh]
[--ff_rnndn_xiph]
[--ff_fftdn [0 - 97]]
[--ff_tempo [0.5 - 2.0]]
[--ff_gate]
[--ff_speechnorm]
[--ff_loudnorm]
[--ff_silence_suppress noise duration]
[--ff_lowhighpass]
</value>
</data>
<data name="textBoxCpp.Text" xml:space="preserve">
<value> -t N, --threads N [4 ] number of threads to use during computation
-p N, --processors N [1 ] number of processors to use during computation
@ -235,86 +315,6 @@
(requires --word_timestamps True) the maximum number of lines in a segment (default: None)
--threads THREADS number of threads used by torch for CPU inference; supercedes MKL_NUM_THREADS/OMP_NUM_THREADS
(default: 0)
</value>
</data>
<data name="textBoxPurfviewFasterWhisper.Text" xml:space="preserve">
<value>[--device DEVICE]
[--output_dir OUTPUT_DIR]
[--output_format {lrc,txt,text,vtt,srt,tsv,json,all}]
[--verbose VERBOSE]
[--task {transcribe,translate}]
[--temperature TEMPERATURE]
[--best_of BEST_OF]
[--beam_size BEAM_SIZE]
[--patience PATIENCE]
[--length_penalty LENGTH_PENALTY]
[--repetition_penalty REPETITION_PENALTY]
[--no_repeat_ngram_size NO_REPEAT_NGRAM_SIZE]
[--suppress_blank SUPPRESS_BLANK]
[--suppress_tokens SUPPRESS_TOKENS]
[--initial_prompt INITIAL_PROMPT]
[--prefix PREFIX]
[--condition_on_previous_text CONDITION_ON_PREVIOUS_TEXT]
[--prompt_reset_on_temperature PROMPT_RESET_ON_TEMPERATURE]
[--without_timestamps WITHOUT_TIMESTAMPS]
[--max_initial_timestamp MAX_INITIAL_TIMESTAMP]
[--temperature_increment_on_fallback TEMPERATURE_INCREMENT_ON_FALLBACK]
[--compression_ratio_threshold COMPRESSION_RATIO_THRESHOLD]
[--logprob_threshold LOGPROB_THRESHOLD]
[--no_speech_threshold NO_SPEECH_THRESHOLD]
[--hallucination_silence_threshold HALLUCINATION_SILENCE_THRESHOLD]
[--clip_timestamps CLIP_TIMESTAMPS]
[--no_speech_strict_lvl {0,1,2}]
[--word_timestamps WORD_TIMESTAMPS]
[--highlight_words HIGHLIGHT_WORDS]
[--prepend_punctuations PREPEND_PUNCTUATIONS]
[--append_punctuations APPEND_PUNCTUATIONS]
[--threads THREADS]
[--version]
[--vad_filter VAD_FILTER]
[--vad_threshold VAD_THRESHOLD]
[--vad_min_speech_duration_ms VAD_MIN_SPEECH_DURATION_MS]
[--vad_max_speech_duration_s VAD_MAX_SPEECH_DURATION_S]
[--vad_min_silence_duration_ms VAD_MIN_SILENCE_DURATION_MS]
[--vad_speech_pad_ms VAD_SPEECH_PAD_MS]
[--vad_window_size_samples VAD_WINDOW_SIZE_SAMPLES]
[--max_new_tokens MAX_NEW_TOKENS]
[--chunk_length CHUNK_LENGTH]
[--compute_type {default,auto,int8,int8_float16,int8_float32,int8_bfloat16,int16,float16,float32,bfloat16}]
[--batch_recursive]
[--beep_off]
[--skip]
[--checkcuda]
[--print_progress]
[--postfix]
[--check_files]
[--PR163_off]
[--hallucinations_list_off]
[--one_word {0,1,2}]
[--sentence]
[--standard]
[--standard_asia]
[--max_comma MAX_COMMA]
[--max_comma_cent {50,60,70,80,90,100}]
[--max_gap MAX_GAP]
[--max_line_width MAX_LINE_WIDTH]
[--max_line_count MAX_LINE_COUNT]
[--min_dist_to_end {0,4,5,6,7,8,9,10,11,12}]
[--prompt_max {16,32,64,128,223}]
[--reprompt {0,1,2}]
[--prompt_reset_on_no_end {0,1,2}]
[--ff_dump]
[--ff_mp3]
[--ff_sync]
[--ff_rnndn_sh]
[--ff_rnndn_xiph]
[--ff_fftdn [0 - 97]]
[--ff_tempo [0.5 - 2.0]]
[--ff_gate]
[--ff_speechnorm]
[--ff_loudnorm]
[--ff_silence_suppress noise duration]
[--ff_lowhighpass]
</value>
</data>
</root>