Update app.py
Browse files
app.py
CHANGED
@@ -113,8 +113,16 @@ class App:
|
|
113 |
with gr.Accordion("Speaker diarization", open=False, visible=True):
|
114 |
cb_diarize = gr.Checkbox(value=diarization_params["is_diarize"],label="Use diarization",interactive=True)
|
115 |
tb_hf_token = gr.Text(label="Token", value=diarization_params["hf_token"],info="An access token is required to use diarization & can be created [here](https://hf.co/settings/tokens). If not done yet for your account, you need to accept the terms & conditions of [diarization](https://huggingface.co/pyannote/speaker-diarization-3.1) & [segmentation](https://huggingface.co/pyannote/segmentation-3.0)")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
116 |
|
117 |
-
with gr.Accordion("Advanced options", open=False, visible=
|
118 |
with gr.Accordion("Advanced diarization options", open=False, visible=True):
|
119 |
dd_diarization_device = gr.Dropdown(label="Device",
|
120 |
choices=self.whisper_inf.diarizer.get_available_device(),
|
@@ -192,10 +200,9 @@ class App:
|
|
192 |
nb_batch_size = gr.Number(label="Batch Size", value=whisper_params["batch_size"], precision=0)
|
193 |
|
194 |
with gr.Accordion("Background Music Remover Filter", open=False):
|
195 |
-
cb_bgm_separation = gr.Checkbox(label="Enable Background Music Remover Filter", value=uvr_params["is_separate_bgm"],
|
196 |
-
interactive=True,
|
197 |
-
info="Enabling this will remove background music by submodel before"
|
198 |
-
" transcribing ")
|
199 |
dd_uvr_device = gr.Dropdown(label="Device", value=self.whisper_inf.music_separator.device,
|
200 |
choices=self.whisper_inf.music_separator.available_devices)
|
201 |
dd_uvr_model_size = gr.Dropdown(label="Model", value=uvr_params["model_size"],
|
@@ -206,9 +213,9 @@ class App:
|
|
206 |
value=uvr_params["enable_offload"])
|
207 |
|
208 |
with gr.Accordion("Voice Detection Filter", open=False):
|
209 |
-
cb_vad_filter = gr.Checkbox(label="Enable Silero VAD Filter", value=vad_params["vad_filter"],
|
210 |
-
interactive=True,
|
211 |
-
info="Enable this to transcribe only detected voice parts by submodel.")
|
212 |
sd_threshold = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, label="Speech Threshold",
|
213 |
value=vad_params["threshold"],
|
214 |
info="Lower it to be more sensitive to small sounds.")
|
|
|
113 |
with gr.Accordion("Speaker diarization", open=False, visible=True):
|
114 |
cb_diarize = gr.Checkbox(value=diarization_params["is_diarize"],label="Use diarization",interactive=True)
|
115 |
tb_hf_token = gr.Text(label="Token", value=diarization_params["hf_token"],info="An access token is required to use diarization & can be created [here](https://hf.co/settings/tokens). If not done yet for your account, you need to accept the terms & conditions of [diarization](https://huggingface.co/pyannote/speaker-diarization-3.1) & [segmentation](https://huggingface.co/pyannote/segmentation-3.0)")
|
116 |
+
|
117 |
+
with gr.Accordion("Advanced audio options", open=False, visible=True):
|
118 |
+
cb_bgm_separation = gr.Checkbox(label="Enable Background Music Remover Filter", value=uvr_params["is_separate_bgm"],
|
119 |
+
interactive=True,
|
120 |
+
info="Enable to remove background music before transcribing")
|
121 |
+
cb_vad_filter = gr.Checkbox(label="Enable Silero VAD Filter", value=vad_params["vad_filter"],
|
122 |
+
interactive=True,
|
123 |
+
info="Enable to transcribe only detected voice parts")
|
124 |
|
125 |
+
with gr.Accordion("Advanced options", open=False, visible=False):
|
126 |
with gr.Accordion("Advanced diarization options", open=False, visible=True):
|
127 |
dd_diarization_device = gr.Dropdown(label="Device",
|
128 |
choices=self.whisper_inf.diarizer.get_available_device(),
|
|
|
200 |
nb_batch_size = gr.Number(label="Batch Size", value=whisper_params["batch_size"], precision=0)
|
201 |
|
202 |
with gr.Accordion("Background Music Remover Filter", open=False):
|
203 |
+
# cb_bgm_separation = gr.Checkbox(label="Enable Background Music Remover Filter", value=uvr_params["is_separate_bgm"],
|
204 |
+
# interactive=True,
|
205 |
+
# info="Enabling this will remove background music by submodel before transcribing.")
|
|
|
206 |
dd_uvr_device = gr.Dropdown(label="Device", value=self.whisper_inf.music_separator.device,
|
207 |
choices=self.whisper_inf.music_separator.available_devices)
|
208 |
dd_uvr_model_size = gr.Dropdown(label="Model", value=uvr_params["model_size"],
|
|
|
213 |
value=uvr_params["enable_offload"])
|
214 |
|
215 |
with gr.Accordion("Voice Detection Filter", open=False):
|
216 |
+
# cb_vad_filter = gr.Checkbox(label="Enable Silero VAD Filter", value=vad_params["vad_filter"],
|
217 |
+
# interactive=True,
|
218 |
+
# info="Enable this to transcribe only detected voice parts by submodel.")
|
219 |
sd_threshold = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, label="Speech Threshold",
|
220 |
value=vad_params["threshold"],
|
221 |
info="Lower it to be more sensitive to small sounds.")
|