Flux9665 commited on
Commit
8e89766
β€’
1 Parent(s): 83f461d

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +3 -11
app.py CHANGED
@@ -1,10 +1,5 @@
1
  import os
2
 
3
- import gradio as gr
4
- import numpy as np
5
- import soundfile as sf
6
- import torch
7
-
8
  os.system("git clone --branch v2.3 https://github.com/DigitalPhonetics/IMS-Toucan.git toucan_codebase")
9
  os.system("mv toucan_codebase/* .")
10
 
@@ -12,18 +7,15 @@ from run_model_downloader import download_models
12
 
13
  download_models()
14
 
15
-
16
-
17
  import gradio as gr
18
  import numpy as np
19
  import torch
20
- import math
21
- import os
22
  from Preprocessing.TextFrontend import ArticulatoryCombinedTextFrontend
23
  from TrainingInterfaces.Text_to_Spectrogram.AutoAligner.Aligner import Aligner
24
  from TrainingInterfaces.Text_to_Spectrogram.FastSpeech2.DurationCalculator import DurationCalculator
25
  from InferenceInterfaces.UtteranceCloner import UtteranceCloner
26
 
 
27
  def float2pcm(sig, dtype='int16'):
28
  """
29
  https://gist.github.com/HudsonHuang/fbdf8e9af7993fe2a91620d3fb86a182
@@ -174,7 +166,7 @@ class TTS_Interface:
174
  durations=duration,
175
  pitch=pitch,
176
  energy=self.energy,
177
- phones=True).cpu()
178
 
179
  if self.current_voice == "female":
180
  cloned_speech = self.cloned_speech_female
@@ -207,7 +199,7 @@ iface = gr.Interface(fn=poem_model.read,
207
  gr.outputs.Audio(type="numpy", label="Poem cloned from a reference"),
208
  gr.outputs.Audio(type="numpy", label="Poem after human-in-the-loop adjustments")],
209
  layout="vertical",
210
- title="PoeticTTS - Customizing Poetry for Literary Studies",
211
  thumbnail="Utility/toucan.png",
212
  theme="default",
213
  allow_flagging="never",
 
1
  import os
2
 
 
 
 
 
 
3
  os.system("git clone --branch v2.3 https://github.com/DigitalPhonetics/IMS-Toucan.git toucan_codebase")
4
  os.system("mv toucan_codebase/* .")
5
 
 
7
 
8
  download_models()
9
 
 
 
10
  import gradio as gr
11
  import numpy as np
12
  import torch
 
 
13
  from Preprocessing.TextFrontend import ArticulatoryCombinedTextFrontend
14
  from TrainingInterfaces.Text_to_Spectrogram.AutoAligner.Aligner import Aligner
15
  from TrainingInterfaces.Text_to_Spectrogram.FastSpeech2.DurationCalculator import DurationCalculator
16
  from InferenceInterfaces.UtteranceCloner import UtteranceCloner
17
 
18
+
19
  def float2pcm(sig, dtype='int16'):
20
  """
21
  https://gist.github.com/HudsonHuang/fbdf8e9af7993fe2a91620d3fb86a182
 
166
  durations=duration,
167
  pitch=pitch,
168
  energy=self.energy,
169
+ input_is_phones=True).cpu()
170
 
171
  if self.current_voice == "female":
172
  cloned_speech = self.cloned_speech_female
 
199
  gr.outputs.Audio(type="numpy", label="Poem cloned from a reference"),
200
  gr.outputs.Audio(type="numpy", label="Poem after human-in-the-loop adjustments")],
201
  layout="vertical",
202
+ title="PoeticTTS - Customizing Poetry",
203
  thumbnail="Utility/toucan.png",
204
  theme="default",
205
  allow_flagging="never",