Optional spectrogram (#519)

* use gr.Group more * make spectrogram optional
kahst · Dec 2, 2024 · 0e5aca0 · 0e5aca0
1 parent cdb4d39
commit 0e5aca0
Show file tree

Hide file tree

Showing 11 changed files with 126 additions and 99 deletions.
diff --git a/birdnet_analyzer/gui/multi_file.py b/birdnet_analyzer/gui/multi_file.py
@@ -140,28 +140,29 @@ def select_directory_wrapper():  # Nishant - Function modified for For Folder se
         ) = gu.species_lists()
 
         with gr.Accordion(loc.localize("multi-tab-output-accordion-label"), open=True):
-            output_type_radio = gr.CheckboxGroup(
-                list(OUTPUT_TYPE_MAP.items()),
-                value="table",
-                label=loc.localize("multi-tab-output-radio-label"),
-                info=loc.localize("multi-tab-output-radio-info"),
-            )
+            with gr.Group():
+                output_type_radio = gr.CheckboxGroup(
+                    list(OUTPUT_TYPE_MAP.items()),
+                    value="table",
+                    label=loc.localize("multi-tab-output-radio-label"),
+                    info=loc.localize("multi-tab-output-radio-info"),
+                )
+
+                with gr.Row():
+                    with gr.Column():
+                        combine_tables_checkbox = gr.Checkbox(
+                            False,
+                            label=loc.localize("multi-tab-output-combine-tables-checkbox-label"),
+                            info=loc.localize("multi-tab-output-combine-tables-checkbox-info"),
+                        )
 
-            with gr.Row():
-                with gr.Column():
-                    combine_tables_checkbox = gr.Checkbox(
+                with gr.Row():
+                    skip_existing_checkbox = gr.Checkbox(
                         False,
-                        label=loc.localize("multi-tab-output-combine-tables-checkbox-label"),
-                        info=loc.localize("multi-tab-output-combine-tables-checkbox-info"),
+                        label=loc.localize("multi-tab-skip-existing-checkbox-label"),
+                        info=loc.localize("multi-tab-skip-existing-checkbox-info"),
                     )
 
-            with gr.Row():
-                skip_existing_checkbox = gr.Checkbox(
-                    False,
-                    label=loc.localize("multi-tab-skip-existing-checkbox-label"),
-                    info=loc.localize("multi-tab-skip-existing-checkbox-info"),
-                )
-
         with gr.Row():
             batch_size_number = gr.Number(
                 precision=1,

diff --git a/birdnet_analyzer/gui/review.py b/birdnet_analyzer/gui/review.py
@@ -147,8 +147,9 @@ def create_log_plot(positives, negatives, fig_num=None):
                             undo_btn = gr.Button(loc.localize("review-tab-undo-button-label"))
                         positive_btn = gr.Button(loc.localize("review-tab-pos-button-label"))
                         negative_btn = gr.Button(loc.localize("review-tab-neg-button-label"))
-                        review_audio = gr.Audio(type="filepath", sources=[], show_download_button=False, autoplay=True)
-                        autoplay_checkbox = gr.Checkbox(True, label=loc.localize("review-tab-autoplay-checkbox-label"))
+                        with gr.Group():
+                            review_audio = gr.Audio(type="filepath", sources=[], show_download_button=False, autoplay=True)
+                            autoplay_checkbox = gr.Checkbox(True, label=loc.localize("review-tab-autoplay-checkbox-label"))
 
             no_samles_label = gr.Label(loc.localize("review-tab-no-files-label"), visible=False)
             species_regression_plot = gr.Plot(label=loc.localize("review-tab-regression-plot-label"))

diff --git a/birdnet_analyzer/gui/single_file.py b/birdnet_analyzer/gui/single_file.py
@@ -83,7 +83,9 @@ def runSingleFileAnalysis(
 def build_single_analysis_tab():
     with gr.Tab(loc.localize("single-tab-title")):
         audio_input = gr.Audio(type="filepath", label=loc.localize("single-audio-label"), sources=["upload"])
-        spectogram_output = gr.Plot(label=loc.localize("review-tab-spectrogram-plot-label"), visible=False)
+        with gr.Group():
+            spectogram_output = gr.Plot(label=loc.localize("review-tab-spectrogram-plot-label"), visible=False)
+            generate_spectrogram_cb = gr.Checkbox(value=True, label=loc.localize("single-tab-spectrogram-checkbox-label"), info="Potentially slow for long audio files.")
         audio_path_state = gr.State()
 
         confidence_slider, sensitivity_slider, overlap_slider, fmin_number, fmax_number = gu.sample_sliders(False)
@@ -100,19 +102,32 @@ def build_single_analysis_tab():
         ) = gu.species_lists(False)
         locale_radio = gu.locale()
 
-        def get_audio_path(i):
+        def get_audio_path(i, generate_spectrogram):
             if i:
                 return (
                     i["path"],
                     gr.Audio(label=os.path.basename(i["path"])),
-                    gr.Plot(visible=True, value=utils.spectrogram_from_file(i["path"], fig_size="auto")),
+                    gr.Plot(visible=True, value=utils.spectrogram_from_file(i["path"], fig_size="auto")) if generate_spectrogram else gr.Plot(visible=False),
                 )
             else:
                 return None, None, gr.Plot(visible=False)
 
+        def try_generate_spectrogram(audio_path, generate_spectrogram):
+            if audio_path and generate_spectrogram:
+                return gr.Plot(visible=True, value=utils.spectrogram_from_file(audio_path["path"], fig_size="auto"))
+            else:
+                return gr.Plot()
+
+        generate_spectrogram_cb.change(
+            try_generate_spectrogram,
+            inputs=[audio_input, generate_spectrogram_cb],
+            outputs=spectogram_output,
+            preprocess=False,
+        )
+
         audio_input.change(
             get_audio_path,
-            inputs=audio_input,
+            inputs=[audio_input, generate_spectrogram_cb],
             outputs=[audio_path_state, audio_input, spectogram_output],
             preprocess=False,
         )

diff --git a/birdnet_analyzer/gui/utils.py b/birdnet_analyzer/gui/utils.py
@@ -234,46 +234,47 @@ def sample_sliders(opened=True):
         (Slider (min confidence), Slider (sensitivity), Slider (overlap))
     """
     with gr.Accordion(loc.localize("inference-settings-accordion-label"), open=opened):
-        with gr.Row():
-            confidence_slider = gr.Slider(
-                minimum=0,
-                maximum=1,
-                value=0.5,
-                step=0.01,
-                label=loc.localize("inference-settings-confidence-slider-label"),
-                info=loc.localize("inference-settings-confidence-slider-info"),
-            )
-            sensitivity_slider = gr.Slider(
-                minimum=0.5,
-                maximum=1.5,
-                value=1,
-                step=0.01,
-                label=loc.localize("inference-settings-sensitivity-slider-label"),
-                info=loc.localize("inference-settings-sensitivity-slider-info"),
-            )
-            overlap_slider = gr.Slider(
-                minimum=0,
-                maximum=2.99,
-                value=0,
-                step=0.01,
-                label=loc.localize("inference-settings-overlap-slider-label"),
-                info=loc.localize("inference-settings-overlap-slider-info"),
-            )
+        with gr.Group():
+            with gr.Row():
+                confidence_slider = gr.Slider(
+                    minimum=0,
+                    maximum=1,
+                    value=0.5,
+                    step=0.01,
+                    label=loc.localize("inference-settings-confidence-slider-label"),
+                    info=loc.localize("inference-settings-confidence-slider-info"),
+                )
+                sensitivity_slider = gr.Slider(
+                    minimum=0.5,
+                    maximum=1.5,
+                    value=1,
+                    step=0.01,
+                    label=loc.localize("inference-settings-sensitivity-slider-label"),
+                    info=loc.localize("inference-settings-sensitivity-slider-info"),
+                )
+                overlap_slider = gr.Slider(
+                    minimum=0,
+                    maximum=2.99,
+                    value=0,
+                    step=0.01,
+                    label=loc.localize("inference-settings-overlap-slider-label"),
+                    info=loc.localize("inference-settings-overlap-slider-info"),
+                )
 
-        with gr.Row():
-            fmin_number = gr.Number(
-                cfg.SIG_FMIN,
-                minimum=0,
-                label=loc.localize("inference-settings-fmin-number-label"),
-                info=loc.localize("inference-settings-fmin-number-info"),
-            )
+            with gr.Row():
+                fmin_number = gr.Number(
+                    cfg.SIG_FMIN,
+                    minimum=0,
+                    label=loc.localize("inference-settings-fmin-number-label"),
+                    info=loc.localize("inference-settings-fmin-number-info"),
+                )
 
-            fmax_number = gr.Number(
-                cfg.SIG_FMAX,
-                minimum=0,
-                label=loc.localize("inference-settings-fmax-number-label"),
-                info=loc.localize("inference-settings-fmax-number-info"),
-            )
+                fmax_number = gr.Number(
+                    cfg.SIG_FMAX,
+                    minimum=0,
+                    label=loc.localize("inference-settings-fmax-number-label"),
+                    info=loc.localize("inference-settings-fmax-number-info"),
+                )
 
         return confidence_slider, sensitivity_slider, overlap_slider, fmin_number, fmax_number
 
@@ -298,46 +299,48 @@ def locale():
 
 
 def species_list_coordinates():
-    lat_number = gr.Slider(
-        minimum=-90,
-        maximum=90,
-        value=0,
-        step=1,
-        label=loc.localize("species-list-coordinates-lat-number-label"),
-        info=loc.localize("species-list-coordinates-lat-number-info"),
-    )
-    lon_number = gr.Slider(
-        minimum=-180,
-        maximum=180,
-        value=0,
-        step=1,
-        label=loc.localize("species-list-coordinates-lon-number-label"),
-        info=loc.localize("species-list-coordinates-lon-number-info"),
-    )
-    with gr.Row():
-        yearlong_checkbox = gr.Checkbox(True, label=loc.localize("species-list-coordinates-yearlong-checkbox-label"))
-        week_number = gr.Slider(
-            minimum=1,
-            maximum=48,
-            value=1,
+    with gr.Group():
+        lat_number = gr.Slider(
+            minimum=-90,
+            maximum=90,
+            value=0,
             step=1,
-            interactive=False,
-            label=loc.localize("species-list-coordinates-week-slider-label"),
-            info=loc.localize("species-list-coordinates-week-slider-info"),
+            label=loc.localize("species-list-coordinates-lat-number-label"),
+            info=loc.localize("species-list-coordinates-lat-number-info"),
+        )
+        lon_number = gr.Slider(
+            minimum=-180,
+            maximum=180,
+            value=0,
+            step=1,
+            label=loc.localize("species-list-coordinates-lon-number-label"),
+            info=loc.localize("species-list-coordinates-lon-number-info"),
         )
+        with gr.Row():
+            yearlong_checkbox = gr.Checkbox(True, label=loc.localize("species-list-coordinates-yearlong-checkbox-label"))
+            week_number = gr.Slider(
+                minimum=1,
+                maximum=48,
+                value=1,
+                step=1,
+                interactive=False,
+                label=loc.localize("species-list-coordinates-week-slider-label"),
+                info=loc.localize("species-list-coordinates-week-slider-info"),
+            )
 
-        def onChange(use_yearlong):
-            return gr.Slider(interactive=(not use_yearlong))
-
-        yearlong_checkbox.change(onChange, inputs=yearlong_checkbox, outputs=week_number, show_progress=False)
-    sf_thresh_number = gr.Slider(
-        minimum=0.01,
-        maximum=0.99,
-        value=0.03,
-        step=0.01,
-        label=loc.localize("species-list-coordinates-threshold-slider-label"),
-        info=loc.localize("species-list-coordinates-threshold-slider-info"),
-    )
+        sf_thresh_number = gr.Slider(
+            minimum=0.01,
+            maximum=0.99,
+            value=0.03,
+            step=0.01,
+            label=loc.localize("species-list-coordinates-threshold-slider-label"),
+            info=loc.localize("species-list-coordinates-threshold-slider-info"),
+        )
+    def onChange(use_yearlong):
+        return gr.Slider(interactive=(not use_yearlong))
+
+    yearlong_checkbox.change(onChange, inputs=yearlong_checkbox, outputs=week_number, show_progress=False)
+
 
     return lat_number, lon_number, week_number, sf_thresh_number, yearlong_checkbox
 

diff --git a/birdnet_analyzer/lang/de.json b/birdnet_analyzer/lang/de.json
@@ -1,6 +1,7 @@
 {
     "single-tab-title": "Einzelanalyse",
     "single-audio-label": "Datei",
+    "single-tab-spectrogram-checkbox-label": "Spektrogramm generieren",
     "single-tab-output-header-start": "Start",
     "single-tab-output-header-end": "Ende",
     "single-tab-output-header-sci-name": "Wissenschaftlicher Name",

diff --git a/birdnet_analyzer/lang/en.json b/birdnet_analyzer/lang/en.json
@@ -1,6 +1,7 @@
 {
     "single-tab-title": "Single analysis",
     "single-audio-label": "file",
+    "single-tab-spectrogram-checkbox-label": "Generate spectrogram",
     "single-tab-output-header-start": "Start",
     "single-tab-output-header-end": "End",
     "single-tab-output-header-sci-name": "Scientific name",

diff --git a/birdnet_analyzer/lang/fr.json b/birdnet_analyzer/lang/fr.json
@@ -1,6 +1,7 @@
 {
     "single-tab-title": "Analyse rapide",
     "single-audio-label": "fichier",
+    "single-tab-spectrogram-checkbox-label": "Générer le spectrogramme",
     "single-tab-output-header-start": "Commencer",
     "single-tab-output-header-end": "Arrêter",
     "single-tab-output-header-sci-name": "Nom scientifique",

diff --git a/birdnet_analyzer/lang/pt-br.json b/birdnet_analyzer/lang/pt-br.json
@@ -1,6 +1,7 @@
 {
     "single-tab-title": "Análise Individual",
     "single-audio-label": "arquivo",
+    "single-tab-spectrogram-checkbox-label": "Gerar espectrograma",
     "single-tab-output-header-start": "Início",
     "single-tab-output-header-end": "Fim",
     "single-tab-output-header-sci-name": "Nome científico",

diff --git a/birdnet_analyzer/lang/ru.json b/birdnet_analyzer/lang/ru.json
@@ -1,6 +1,7 @@
 {
     "single-tab-title": "Одиночный анализ",
     "single-audio-label": "файл",
+    "single-tab-spectrogram-checkbox-label": "Генерирование спектрограммы",
     "single-tab-output-header-start": "Старт (сек)",
     "single-tab-output-header-end": "Конец (сек)",
     "single-tab-output-header-sci-name": "Научное название",

diff --git a/birdnet_analyzer/lang/tlh.json b/birdnet_analyzer/lang/tlh.json
@@ -1,6 +1,7 @@
 {
     "single-tab-title": "wa' poj",
     "single-audio-label": "De' wav",
+    "single-tab-spectrogram-checkbox-label": "nuvmey vItlhutlh",
     "single-tab-output-header-start": "Tagh (s)",
     "single-tab-output-header-end": "Dor (s)",
     "single-tab-output-header-sci-name": "Hol qonwI' pong",

diff --git a/birdnet_analyzer/lang/zh_TW.json b/birdnet_analyzer/lang/zh_TW.json
@@ -1,6 +1,7 @@
 {
     "single-tab-title": "單筆音檔",
     "single-audio-label": "檔案",
+    "single-tab-spectrogram-checkbox-label": "產生頻譜圖",
     "single-tab-output-header-start": "開始",
     "single-tab-output-header-end": "結束",
     "single-tab-output-header-sci-name": "學名",