Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -30,7 +30,7 @@ def text2image_latent(text, steps, width, height, images, diversity):
|
|
30 |
return(image_paths)
|
31 |
|
32 |
|
33 |
-
def speech_to_text(mic=None, file=None):
|
34 |
if mic is not None:
|
35 |
audio = mic
|
36 |
elif file is not None:
|
@@ -38,7 +38,8 @@ def speech_to_text(mic=None, file=None):
|
|
38 |
else:
|
39 |
return "You must either provide a mic recording or a file"
|
40 |
transcription = asr(audio)["text"]
|
41 |
-
|
|
|
42 |
|
43 |
def zero_shot(image, text_input):
|
44 |
PIL_image = Image.fromarray(np.uint8(image)).convert('RGB')
|
@@ -66,7 +67,7 @@ with gr.Blocks() as demo:
|
|
66 |
with gr.Row():
|
67 |
with gr.Column():
|
68 |
audio_file =[
|
69 |
-
gr.Audio(source="microphone", type="filepath", optional=True),
|
70 |
gr.Audio(source="upload", type="filepath", optional=True)]
|
71 |
text = gr.Textbox(label="Text", placeholder="If you dont want to record or upload your voice you can input text here")
|
72 |
with gr.Row():
|
@@ -88,8 +89,8 @@ with gr.Blocks() as demo:
|
|
88 |
zero_shot_clf = gr.Button("Classify Image go brr")
|
89 |
|
90 |
|
91 |
-
speech_to_text.click(speech_to_text, inputs=audio_file, outputs=text)
|
92 |
get_image_latent.click(text2image_latent, inputs=[text, steps, width, height, images, diversity], outputs=gallery)
|
93 |
zero_shot_clf.click(zero_shot, inputs=[gallery, text_input], outputs=label)
|
94 |
|
95 |
-
demo.launch()
|
|
|
30 |
return(image_paths)
|
31 |
|
32 |
|
33 |
+
def speech_to_text(mic=None, file=None, state=""):
|
34 |
if mic is not None:
|
35 |
audio = mic
|
36 |
elif file is not None:
|
|
|
38 |
else:
|
39 |
return "You must either provide a mic recording or a file"
|
40 |
transcription = asr(audio)["text"]
|
41 |
+
state += text + " "
|
42 |
+
return state, state
|
43 |
|
44 |
def zero_shot(image, text_input):
|
45 |
PIL_image = Image.fromarray(np.uint8(image)).convert('RGB')
|
|
|
67 |
with gr.Row():
|
68 |
with gr.Column():
|
69 |
audio_file =[
|
70 |
+
gr.Audio(source="microphone", type="filepath", optional=True), "state",
|
71 |
gr.Audio(source="upload", type="filepath", optional=True)]
|
72 |
text = gr.Textbox(label="Text", placeholder="If you dont want to record or upload your voice you can input text here")
|
73 |
with gr.Row():
|
|
|
89 |
zero_shot_clf = gr.Button("Classify Image go brr")
|
90 |
|
91 |
|
92 |
+
speech_to_text.click(speech_to_text, inputs=audio_file, outputs=[text,"state"])
|
93 |
get_image_latent.click(text2image_latent, inputs=[text, steps, width, height, images, diversity], outputs=gallery)
|
94 |
zero_shot_clf.click(zero_shot, inputs=[gallery, text_input], outputs=label)
|
95 |
|
96 |
+
demo.launch(live=True)
|