owaski commited on
Commit
5c51862
·
1 Parent(s): d07cbce

reorgnize the UI

Browse files
Files changed (1) hide show
  1. app.py +39 -42
app.py CHANGED
@@ -527,6 +527,45 @@ with gr.Blocks(
527
  (https://github.com/siddhu001/espnet/tree/sds_demo_recipe/egs2/TEMPLATE/sds1#how-to-use).
528
  """
529
  )
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
530
  with gr.Row():
531
  with gr.Column(scale=1):
532
  user_audio = gr.Audio(
@@ -536,48 +575,6 @@ with gr.Blocks(
536
  format="wav",
537
  )
538
  process_btn = gr.Button("Process Audio", variant="primary")
539
- input_text=gr.Textbox(
540
- label="LLM prompt",
541
- visible=True,
542
- interactive=True,
543
- value=default_instruct
544
- )
545
- with gr.Row():
546
- type_radio = gr.Radio(
547
- choices=["Cascaded"],
548
- label="Choose type of Spoken Dialog:",
549
- value="Cascaded",
550
- )
551
- with gr.Row():
552
- ASR_radio = gr.Radio(
553
- choices=ASR_options,
554
- label="Choose ASR:",
555
- value=ASR_name,
556
- )
557
- with gr.Row():
558
- LLM_radio = gr.Radio(
559
- choices=LLM_options,
560
- label="Choose LLM:",
561
- value=LLM_name,
562
- )
563
- with gr.Row():
564
- radio = gr.Radio(
565
- choices=TTS_display_names,
566
- label="Choose TTS:",
567
- value=TTS_reverse_map[TTS_name],
568
- )
569
- with gr.Row():
570
- E2Eradio = gr.Radio(
571
- choices=["mini-omni"],
572
- label="Choose E2E model:",
573
- value="mini-omni",
574
- visible=False,
575
- )
576
- gr.Examples(
577
- examples=[[row[1], row[2]] for row in examples],
578
- inputs=[input_text, radio],
579
- label="Usage examples",
580
- )
581
  with gr.Column(scale=1):
582
  output_asr_text = gr.Textbox(label="ASR Transcription", interactive=False)
583
  output_text = gr.Textbox(label="LLM Response", interactive=False)
 
527
  (https://github.com/siddhu001/espnet/tree/sds_demo_recipe/egs2/TEMPLATE/sds1#how-to-use).
528
  """
529
  )
530
+ input_text = gr.Textbox(
531
+ label="LLM prompt",
532
+ visible=True,
533
+ interactive=True,
534
+ value=default_instruct,
535
+ )
536
+ with gr.Row():
537
+ type_radio = gr.Radio(
538
+ choices=["Cascaded"],
539
+ label="Choose type of Spoken Dialog:",
540
+ value="Cascaded",
541
+ )
542
+ ASR_radio = gr.Radio(
543
+ choices=ASR_options,
544
+ label="Choose ASR:",
545
+ value=ASR_name,
546
+ )
547
+ LLM_radio = gr.Radio(
548
+ choices=LLM_options,
549
+ label="Choose LLM:",
550
+ value=LLM_name,
551
+ )
552
+ with gr.Column(scale=0.5):
553
+ radio = gr.Radio(
554
+ choices=TTS_display_names,
555
+ label="Choose TTS:",
556
+ value=TTS_reverse_map[TTS_name],
557
+ )
558
+ E2Eradio = gr.Radio(
559
+ choices=["mini-omni"],
560
+ label="Choose E2E model:",
561
+ value="mini-omni",
562
+ visible=False,
563
+ )
564
+ gr.Examples(
565
+ examples=[[row[1], row[2]] for row in examples],
566
+ inputs=[input_text, radio],
567
+ label="Usage examples",
568
+ )
569
  with gr.Row():
570
  with gr.Column(scale=1):
571
  user_audio = gr.Audio(
 
575
  format="wav",
576
  )
577
  process_btn = gr.Button("Process Audio", variant="primary")
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
578
  with gr.Column(scale=1):
579
  output_asr_text = gr.Textbox(label="ASR Transcription", interactive=False)
580
  output_text = gr.Textbox(label="LLM Response", interactive=False)