Spaces:
Running
Running
Add share button
Browse files
app.py
CHANGED
@@ -79,14 +79,21 @@ def gradio_generate(prompt, steps, guidance):
|
|
79 |
|
80 |
return output_filename
|
81 |
|
82 |
-
description_text = "
|
83 |
-
|
|
|
|
|
84 |
samples related to concepts that it has not seen in training (e.g. singing). For the same reason, TANGO \
|
85 |
is not always able to finely control its generations over textual control prompts. For example, \
|
86 |
the generations from TANGO for prompts Chopping tomatoes on a wooden table and Chopping potatoes \
|
87 |
on a metal table are very similar. \
|
88 |
-
|
89 |
-
compositional and controllable generation ability.
|
|
|
|
|
|
|
|
|
|
|
90 |
|
91 |
# Gradio input and output components
|
92 |
input_text = gr.inputs.Textbox(lines=2, label="Prompt")
|
|
|
79 |
|
80 |
return output_filename
|
81 |
|
82 |
+
description_text = """
|
83 |
+
<p><a href="https://huggingface.co/spaces/declare-lab/tango/blob/main/app.py?duplicate=true"> <img style="margin-top: 0em; margin-bottom: 0em" src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a> For faster inference without waiting in queue, you may duplicate the space and upgrade to a GPU in the settings. <br/><br/>
|
84 |
+
Generate audio using TANGO by providing a text prompt.
|
85 |
+
<br/><br/>Limitations: TANGO is trained on the small AudioCaps dataset so it may not generate good audio \
|
86 |
samples related to concepts that it has not seen in training (e.g. singing). For the same reason, TANGO \
|
87 |
is not always able to finely control its generations over textual control prompts. For example, \
|
88 |
the generations from TANGO for prompts Chopping tomatoes on a wooden table and Chopping potatoes \
|
89 |
on a metal table are very similar. \
|
90 |
+
<br/><br/>We are currently training another version of TANGO on larger datasets to enhance its generalization, \
|
91 |
+
compositional and controllable generation ability.
|
92 |
+
|
93 |
+
We recommend using a guidance scale of 3. The default number of steps is set to 100. More steps generally lead \
|
94 |
+
to better quality of generated audios but will take a longer time.
|
95 |
+
<p/>
|
96 |
+
"""
|
97 |
|
98 |
# Gradio input and output components
|
99 |
input_text = gr.inputs.Textbox(lines=2, label="Prompt")
|