Muhammed-ai commited on
Commit
8090614
·
verified ·
1 Parent(s): ea51a26

Add sherpa piper voice model archives

Browse files
README.md CHANGED
@@ -1,3 +1,72 @@
1
  ---
2
- license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ license: other
3
+ language:
4
+ - tr
5
+ - en
6
+ tags:
7
+ - text-to-speech
8
+ - tts
9
+ - piper
10
+ - sherpa-onnx
11
+ - onnx
12
+ - mobile
13
+ pipeline_tag: text-to-speech
14
  ---
15
+
16
+ # SES Mobil Piper TTS Models
17
+
18
+ This repository contains Piper VITS text-to-speech voices packaged for sherpa-onnx mobile usage.
19
+
20
+ ## Files
21
+
22
+ | Archive | Language | Voice | Quality | Flutter archiveRoot | Flutter modelFileName |
23
+ | --- | --- | --- | --- | --- | --- |
24
+ | `vits-piper-en_US-bryce-medium.tar.bz2` | English | bryce | medium | `vits-piper-en_US-bryce-medium` | `en_US-bryce-medium.onnx` |
25
+ | `vits-piper-tr_TR-fahrettin-medium.tar.bz2` | Turkish | fahrettin | medium | `vits-piper-tr_TR-fahrettin-medium` | `tr_TR-fahrettin-medium.onnx` |
26
+ | `vits-piper-tr_TR-youtube1-training_data_yeni.tar.bz2` | Turkish | youtube1 | training_data_yeni | `vits-piper-tr_TR-youtube1-training_data_yeni` | `tr_TR-youtube1-training_data_yeni.onnx` |
27
+
28
+ Each archive contains:
29
+
30
+ ```text
31
+ vits-piper-.../
32
+ *.onnx
33
+ tokens.txt
34
+ espeak-ng-data/
35
+ ```
36
+
37
+ The ONNX files include sherpa-onnx metadata:
38
+
39
+ ```text
40
+ model_type: vits
41
+ comment: piper
42
+ has_espeak: 1
43
+ n_speakers: 1
44
+ sample_rate: 22050
45
+ ```
46
+
47
+ ## Flutter Usage
48
+
49
+ Use the Hugging Face direct download URL format:
50
+
51
+ ```text
52
+ https://huggingface.co/Muhammed-ai/TTS/resolve/main/<archive-name>.tar.bz2
53
+ ```
54
+
55
+ Example:
56
+
57
+ ```dart
58
+ VoiceModel(
59
+ id: 'tr_TR-fahrettin-medium',
60
+ displayName: 'Fahrettin Turkce Ton',
61
+ languageLabel: 'Turkce',
62
+ quality: 'Medium',
63
+ archiveUrl:
64
+ 'https://huggingface.co/Muhammed-ai/TTS/resolve/main/vits-piper-tr_TR-fahrettin-medium.tar.bz2',
65
+ archiveRoot: 'vits-piper-tr_TR-fahrettin-medium',
66
+ modelFileName: 'tr_TR-fahrettin-medium.onnx',
67
+ ),
68
+ ```
69
+
70
+ ## Notes
71
+
72
+ These models are prepared for applications that load Piper VITS models through sherpa-onnx `OfflineTtsVitsModelConfig`.
_cache/espeak-ng-data.tar.bz2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4135ccf82e1f40613491c0874d4945ae9e9c7840933d8e25a6f9e003d9ebf533
3
+ size 7252012
manifest.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "source": "en_US-bryce-medium.onnx",
4
+ "archive_root": "vits-piper-en_US-bryce-medium",
5
+ "model_file_name": "en_US-bryce-medium.onnx",
6
+ "archive_path": "C:\\Users\\MUHAMMED\\Desktop\\SES_MOBİL\\Model_dönüşüm\\output\\vits-piper-en_US-bryce-medium.tar.bz2",
7
+ "token_count": 157,
8
+ "metadata": {
9
+ "model_type": "vits",
10
+ "comment": "piper",
11
+ "language": "English",
12
+ "voice": "en",
13
+ "has_espeak": 1,
14
+ "n_speakers": 1,
15
+ "sample_rate": 22050
16
+ }
17
+ },
18
+ {
19
+ "source": "fahrettin.onnx",
20
+ "archive_root": "vits-piper-tr_TR-fahrettin-medium",
21
+ "model_file_name": "tr_TR-fahrettin-medium.onnx",
22
+ "archive_path": "C:\\Users\\MUHAMMED\\Desktop\\SES_MOBİL\\Model_dönüşüm\\output\\vits-piper-tr_TR-fahrettin-medium.tar.bz2",
23
+ "token_count": 159,
24
+ "metadata": {
25
+ "model_type": "vits",
26
+ "comment": "piper",
27
+ "language": "Turkish",
28
+ "voice": "tr",
29
+ "has_espeak": 1,
30
+ "n_speakers": 1,
31
+ "sample_rate": 22050
32
+ }
33
+ },
34
+ {
35
+ "source": "youtube1.onnx",
36
+ "archive_root": "vits-piper-tr_TR-youtube1-training_data_yeni",
37
+ "model_file_name": "tr_TR-youtube1-training_data_yeni.onnx",
38
+ "archive_path": "C:\\Users\\MUHAMMED\\Desktop\\SES_MOBİL\\Model_dönüşüm\\output\\vits-piper-tr_TR-youtube1-training_data_yeni.tar.bz2",
39
+ "token_count": 157,
40
+ "metadata": {
41
+ "model_type": "vits",
42
+ "comment": "piper",
43
+ "language": "Turkish",
44
+ "voice": "tr",
45
+ "has_espeak": 1,
46
+ "n_speakers": 1,
47
+ "sample_rate": 22050
48
+ }
49
+ }
50
+ ]
vits-piper-en_US-bryce-medium.tar.bz2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d1653db48ff067ca467537b71caa68199a4ee949a6f4777829c92dbc41f5471
3
+ size 67276880
vits-piper-tr_TR-fahrettin-medium.tar.bz2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6e1d2938105ba3761e21d011e9f948f764ffff268737f7ce1fc505d3afeba21
3
+ size 67178661
vits-piper-tr_TR-youtube1-training_data_yeni.tar.bz2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bb4c1575cd24a526edccb8d6f3969318d299801235947529f2dce9b9b40f863
3
+ size 67200835