Update example table front-end and logic, adding cache example

app.py

@@ -1,6 +1,7 @@
 import gradio as gr
 import numpy as np
 import librosa
+import os
 
 from asr.run_asr import run_asr_inference, load_asr_model
 from nlu.run_nlu import run_nlu_inference, load_nlu_model
@@ -18,6 +19,8 @@ description=[
     f"""For more details on the implementation, check our {blog_post_link}.""",
 ]
 
+CACHE_EXAMPLES = os.getenv("CACHE_EXAMPLES") == "1"
+
 examples = [
     "resources/audios/utt_286.wav",
     "resources/audios/utt_2414.wav",
@@ -59,6 +62,7 @@ slots = [
     [ 'Other', 'Other', 'Other', 'Other', 'Other', 'Other', 'Other', 'media_type', 'media_type', 'media_type']
 ]
 
+example_list = [[example, transcription, slot, intent] for example, transcription, slot, intent in zip(examples, transcriptions, slots, intents)]
 
 utter_ack_text = """This is an output of the European Project UTTER (Unified Transcription and Translation for Extended Reality) funded by European Union’s Horizon Europe Research and Innovation programme under grant agreement number 101070631. 
 For more information please visit https://he-utter.eu/"""
@@ -71,19 +75,6 @@ nle_logo = """<a href="https://europe.naverlabs.com/" target="_blank"><img src="
 fbk_logo = """<a href="https://mt.fbk.eu/" target="_blank"><img src="https://huggingface.co/spaces/naver/French-SLU-DEMO-Interspeech2024/resolve/main/resources/logos/FBK_logo.png" width="100" height="100"></a>"""
 
 
-table = f"""
-    | File | Transcription | Slots | Intent | 
-    | ------------ | ------------------- | ---------- | -----------| 
-    | {examples[0].split("/")[-1]}  | {transcriptions[0]} | {slots[0]} | {intents[0]} |
-    | {examples[1].split("/")[-1]}  | {transcriptions[1]} | {slots[1]} | {intents[1]} |
-    | {examples[2].split("/")[-1]}  | {transcriptions[2]} | {slots[2]} | {intents[2]} |
-    | {examples[3].split("/")[-1]}  | {transcriptions[3]} | {slots[3]} | {intents[3]} |
-    | {examples[4].split("/")[-1]}  | {transcriptions[4]} | {slots[4]} | {intents[4]} |
-    | {examples[5].split("/")[-1]}  | {transcriptions[5]} | {slots[5]} | {intents[5]} |
-    | {examples[6].split("/")[-1]}  | {transcriptions[6]} | {slots[6]} | {intents[6]} |
-    | {examples[7].split("/")[-1]}  | {transcriptions[7]} | {slots[7]} | {intents[7]} |
-    """.strip()
-
 ############### calls
 
 def run_inference(audio_file):
@@ -126,6 +117,10 @@ with demo:
             show_share_button=False,
             max_length=20,
             )
+    with gr.Row(visible=False):
+        _transcription = gr.Textbox(label="Transcription")
+        _slot = gr.Textbox(label="Slots")
+        _intent = gr.Textbox(label="Intent")
         
     output = gr.HighlightedText(label="ASR result + NLU result")
 
@@ -137,8 +132,12 @@ with demo:
     )
 
     with gr.Row():
-        gr.Examples(label="Speech-MASSIVE test utterances:", inputs=audio_file, examples=examples)
-    gr.Markdown(table)
+        gr.Examples(
+            label="Examples(Speech-MASSIVE test utterances):",
+            examples=example_list,
+            inputs=[audio_file, _transcription, _slot, _intent],
+            cache_examples=CACHE_EXAMPLES,
+        )
 
     gr.Markdown("# Aknowledgments")
     gr.Markdown(utter_ack_text)