[Enhancement] Add waveform output (#5)

- [Enhancement] Add waveform output (2d3ebd5088c5a51c240b5bbcfd9bbc210aac98ca)


Co-authored-by: Jahangir <[email protected]>

app.py

@@ -12,6 +12,7 @@ import json
 import os
 import re
 import os
+import random
 
 import pandas as pd
 
@@ -51,7 +52,6 @@ df_init = pd.DataFrame(columns=['Task', 'Top 1', 'Top 2', 'Top 3', 'Top 4', 'Top
 transcription_df = gr.DataFrame(value=df_init, label="Model Results", row_count=(
     0, "dynamic"), max_rows=30, wrap=True, overflow_row_behaviour='paginate')
 # outputs = [gr.components.Textbox()]
-outputs = transcription_df
 
 df_init_live = pd.DataFrame(columns=['Task', 'Top 1', 'Top 2', 'Top 3', 'Top 4', 'Top 5'])
 transcription_df_live = gr.DataFrame(value=df_init_live, label="Model Results", row_count=(
@@ -187,11 +187,22 @@ def model_infernce(inputs):
     df = pd.DataFrame(df_objects, columns=['Task', 'Top 1', 'Top 2', 'Top 3', 'Top 4', 'Top 5'])
     return df
 
-def convert_audio(inputs, microphone):
-    if (microphone is not None):
-        inputs = microphone
-    df = model_infernce(inputs)
-    return df
+COLORS = [
+    ["#ff0000", "#00ff00"],
+    ["#00ff00", "#0000ff"],
+    ["#0000ff", "#ff0000"],
+]
+
+def convert_audio(audio):
+    df = model_infernce(audio)
+    waveform = gr.make_waveform(audio,
+        bars_color=random.choice(COLORS),
+        bar_count=80,
+        bar_width=0.4)
+    return df, waveform
+
+def audio_waveform(audio):
+    return gr.make_waveform(audio)
 
 def live_convert_audio(microphone):
     if (microphone is not None):
@@ -199,6 +210,8 @@ def live_convert_audio(microphone):
     df = model_infernce(inputs)
     return df
 
+outputs = [transcription_df, gr.Video(label="Waveform")]
+
 audio_chunked = gr.Interface(
     fn=convert_audio,
     inputs=inputs,