Spaces:
Runtime error
Runtime error
add a hidden audio component output
#55
by
fffiloni
- opened
app.py
CHANGED
@@ -26,11 +26,12 @@ audioldm = build_model()
|
|
26 |
def text2audio(text, duration, guidance_scale, random_seed, n_candidates):
|
27 |
# print(text, length, guidance_scale)
|
28 |
waveform = text_to_audio(audioldm, text, random_seed, duration=duration, guidance_scale=guidance_scale, n_candidate_gen_per_text=int(n_candidates)) # [bs, 1, samples]
|
|
|
29 |
waveform = [gr.make_waveform((16000, wave[0]), bg_image="bg.png") for wave in waveform]
|
30 |
# waveform = [(16000, np.random.randn(16000)), (16000, np.random.randn(16000))]
|
31 |
if(len(waveform) == 1):
|
32 |
waveform = waveform[0]
|
33 |
-
return waveform
|
34 |
|
35 |
# iface = gr.Interface(fn=text2audio, inputs=[
|
36 |
# gr.Textbox(value="A man is speaking in a huge room", max_lines=1),
|
@@ -225,7 +226,7 @@ with iface:
|
|
225 |
n_candidates = gr.Slider(1, 5, value=3, step=1, label="Automatic quality control. This number control the number of candidates (e.g., generate three audios and choose the best to show you). A Larger value usually lead to better quality with heavier computation")
|
226 |
############# Output
|
227 |
# outputs=gr.Audio(label="Output", type="numpy")
|
228 |
-
outputs=gr.Video(label="Output", elem_id="output-video")
|
229 |
|
230 |
# with gr.Group(elem_id="container-advanced-btns"):
|
231 |
# # advanced_button = gr.Button("Advanced options", elem_id="advanced-btn")
|
|
|
26 |
def text2audio(text, duration, guidance_scale, random_seed, n_candidates):
|
27 |
# print(text, length, guidance_scale)
|
28 |
waveform = text_to_audio(audioldm, text, random_seed, duration=duration, guidance_scale=guidance_scale, n_candidate_gen_per_text=int(n_candidates)) # [bs, 1, samples]
|
29 |
+
audio_out = waveform
|
30 |
waveform = [gr.make_waveform((16000, wave[0]), bg_image="bg.png") for wave in waveform]
|
31 |
# waveform = [(16000, np.random.randn(16000)), (16000, np.random.randn(16000))]
|
32 |
if(len(waveform) == 1):
|
33 |
waveform = waveform[0]
|
34 |
+
return waveform, audio_out
|
35 |
|
36 |
# iface = gr.Interface(fn=text2audio, inputs=[
|
37 |
# gr.Textbox(value="A man is speaking in a huge room", max_lines=1),
|
|
|
226 |
n_candidates = gr.Slider(1, 5, value=3, step=1, label="Automatic quality control. This number control the number of candidates (e.g., generate three audios and choose the best to show you). A Larger value usually lead to better quality with heavier computation")
|
227 |
############# Output
|
228 |
# outputs=gr.Audio(label="Output", type="numpy")
|
229 |
+
outputs=gr.Video(label="Output", elem_id="output-video"), gr.Audio(visible=False)
|
230 |
|
231 |
# with gr.Group(elem_id="container-advanced-btns"):
|
232 |
# # advanced_button = gr.Button("Advanced options", elem_id="advanced-btn")
|