Update app.py

app.py

@@ -77,11 +77,10 @@ se_demo = gr.Interface(
         gr.Audio(label="Output Audio", type="filepath"),
     ],
     title = "ClearVoice: Speech Enhancement",
-    description = ("Gradio demo for Speech enhancement with ClearVoice. The models support audios with 16 kHz (FRCRN backbone) and 48 kHz (MossFormer2 backbone) sampling rates. "
-                   "We provide the generalized models trained on large scale of data for handling various of background environments. "
-                   "To test it, simply upload your audio, or click one of the examples to load them. Read more at the links below."),
+    description = ("ClearVoice is AI-powered and extracts clear speech from background noise for enhanced speech quality. It supports both 16 kHz and 48 kHz audio outputs. "
+                   "To try it, simply upload your audio, or click one of the examples. "),
     article = ("<p style='text-align: center'><a href='https://arxiv.org/abs/2206.07293' target='_blank'>FRCRN: Boosting Feature Representation Using Frequency Recurrence for Monaural Speech Enhancement</a> | <a href='https://github.com/alibabasglab/FRCRN' target='_blank'>Github Repo</a></p>"
-              ),
+              "<p style='text-align: center'><a href='https://arxiv.org/abs/2312.11825' target='_blank'>MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation</a> | <a href='https://github.com/alibabasglab/MossFormer2' target='_blank'>Github Repo</a></p>"),
     examples = [
         ["examples/mandarin_speech_16kHz.wav", "16000"],
         ["examples/english_speech_48kHz.wav", "48000"],
@@ -99,9 +98,8 @@ ss_demo = gr.Interface(
         gr.Audio(label="Output Audio", type="filepath"),
     ],
     title = "ClearVoice: Speech Separation",
-    description = ("Gradio demo for Speech separation with ClearVoice. The model (MossFormer2 backbone) supports 2 speakers' audio mixtures with 16 kHz sampling rate. "
-                   "We provide the generalized models trained on large scale of data for handling independent speakers and various of background environments. "
-                    "To test it, simply upload your audio, or click one of the examples to load them. Read more at the links below."),
+    description = ("ClearVoice is powered by AI and separates individual speech from mixed audio. It supports 16 kHz and two output streams. 
+                    "To try it, simply upload your audio, or click one of the examples. "),
     article = ("<p style='text-align: center'><a href='https://arxiv.org/abs/2302.11824' target='_blank'>MossFormer: Pushing the Performance Limit of Monaural Speech Separation using Gated Single-Head Transformer with Convolution-Augmented Joint Self-Attentions</a> | <a href='https://github.com/alibabasglab/MossFormer' target='_blank'>Github Repo</a></p>"
               "<p style='text-align: center'><a href='https://arxiv.org/abs/2312.11825' target='_blank'>MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation</a> | <a href='https://github.com/alibabasglab/MossFormer2' target='_blank'>Github Repo</a></p>"),
     examples = [
@@ -119,9 +117,9 @@ tse_demo = gr.Interface(
     outputs = [
         gr.Gallery(label="Output Video List")
     ],
-    title = "ClearVoice: Audio-visual speaker extraction",
-    description = ("Gradio demo for audio-visual speaker extraction with ClearVoice."
-                    "To test it, simply upload your video, or click one of the examples to load them. Read more at the links below."),
+    title = "ClearVoice: Audio-Visual Speaker Extraction",
+    description = ("ClearVoice is AI-powered and extracts each speaker's voice from a multi-speaker video using facial recognition. "
+                    "To try it, simply upload your video, or click one of the examples. "),
     # article = ("<p style='text-align: center'><a href='https://arxiv.org/abs/2302.11824' target='_blank'>MossFormer: Pushing the Performance Limit of Monaural Speech Separation using Gated Single-Head Transformer with Convolution-Augmented Joint Self-Attentions</a> | <a href='https://github.com/alibabasglab/MossFormer' target='_blank'>Github Repo</a></p>"
     #           "<p style='text-align: center'><a href='https://arxiv.org/abs/2312.11825' target='_blank'>MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation</a> | <a href='https://github.com/alibabasglab/MossFormer2' target='_blank'>Github Repo</a></p>"),
     examples = [
@@ -132,7 +130,6 @@ tse_demo = gr.Interface(
 )
 
 with demo:
-    #gr.TabbedInterface([se_demo], ["Speech Enhancement"])
-    gr.TabbedInterface([se_demo, ss_demo, tse_demo], ["Demo_1: Speech Enhancement", "Demo_2: Speech Separation", "Demo_3: Audio-visual Speaker Extraction"])
+    gr.TabbedInterface([se_demo, ss_demo, tse_demo], ["Task 1: Speech Enhancement", "Task 2: Speech Separation", "Task 3: Audio-Visual Speaker Extraction"])
 
 demo.launch()