Readme
Browse files- README.py +31 -0
- tts_harvard.py +2 -2
README.py
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-sa-4.0
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
pipeline_tag: text-to-speech
|
6 |
+
tags:
|
7 |
+
- msp-podcast
|
8 |
+
- emotion-recognition
|
9 |
+
- audio
|
10 |
+
- speech
|
11 |
+
- styletts2
|
12 |
+
- mimic3
|
13 |
+
- speech-emotion-recognition
|
14 |
+
- dkounadis
|
15 |
+
---
|
16 |
+
|
17 |
+
# Artificial StyleTTS2
|
18 |
+
|
19 |
+
Using Mimic-3 Synthetic Speech to Drive StyleTTS2
|
20 |
+
|
21 |
+
|
22 |
+
|
23 |
+
**[arXiv Coming soon]()**
|
24 |
+
|
25 |
+
```
|
26 |
+
|
27 |
+
```
|
28 |
+
|
29 |
+
|
30 |
+
|
31 |
+
See demo at [SHIFT TTS tool](https://github.com/audeering/shift/tree/main)
|
tts_harvard.py
CHANGED
@@ -106,12 +106,12 @@ for audio_prompt in ['english',
|
|
106 |
'human',
|
107 |
'foreign',
|
108 |
'foreign_4x']:
|
109 |
-
OUT_FILE = f'{audio_prompt}
|
110 |
if not os.path.isfile(OUT_FILE):
|
111 |
total_audio = []
|
112 |
total_style = []
|
113 |
ix = 0
|
114 |
-
for list_of_10 in harvard_individual_sentences[:
|
115 |
# long_sentence = ' '.join(list_of_10['sentences'])
|
116 |
# harvard.append(long_sentence.replace('.', ' '))
|
117 |
for text in list_of_10['sentences']:
|
|
|
106 |
'human',
|
107 |
'foreign',
|
108 |
'foreign_4x']:
|
109 |
+
OUT_FILE = f'{audio_prompt}_hfullh.wav'
|
110 |
if not os.path.isfile(OUT_FILE):
|
111 |
total_audio = []
|
112 |
total_style = []
|
113 |
ix = 0
|
114 |
+
for list_of_10 in harvard_individual_sentences[:1000]:
|
115 |
# long_sentence = ' '.join(list_of_10['sentences'])
|
116 |
# harvard.append(long_sentence.replace('.', ' '))
|
117 |
for text in list_of_10['sentences']:
|