ckoch/AIVoices_HA
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add cathy, cynda, and james Piper voices for Home Assistant

Browse Source 
Three medium-quality en-US single-speaker Piper TTS voices, each with
its .onnx model and .onnx.json config, plus README install instructions.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
ckoch
2026-06-12 09:33:42 -04:00
co-authored by Claude Opus 4.8
commit a622e58792
8 changed files with 1564 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.gitignore
+15
View File
@@ -0,0 +1,15 @@
# OS / editor cruft
.DS_Store
Thumbs.db
*~
.idea/
.vscode/
# Python (if you add training/conversion scripts later)
__pycache__/
*.pyc
.venv/
venv/
# Audio test output
*.wav
README.md
+58
View File
@@ -0,0 +1,58 @@
# AIVoices_HA
Custom [Piper](https://github.com/OHF-Voice/piper1-gpl) text-to-speech voices for [Home Assistant](https://www.home-assistant.io/).
These are medium-quality, single-speaker English (`en-US`) neural voices trained for use with Home Assistant's local TTS engine.
## Voices
| Voice | Files | Language | Sample rate | Quality |
|-------|-------|----------|-------------|---------|
| **cathy** | `en_us-cathy-medium.onnx` + `.onnx.json` | en-US | 22050 Hz | medium |
| **cynda** | `en_us-cynda-medium.onnx` + `.onnx.json` | en-US | 22050 Hz | medium |
| **james** | `en_us-james-medium.onnx` + `.onnx.json` | en-US | 22050 Hz | medium |
Each voice consists of two files that must stay together:
- `*.onnx` — the trained model weights
- `*.onnx.json` — the voice configuration (phoneme map, sample rate, inference settings)
## Installation (Home Assistant)
These voices work with the **Piper** add-on / [wyoming-piper](https://github.com/rhasspy/wyoming-piper).
1. Locate the Piper data directory where custom voices are stored. For the
official **Piper** add-on this is the share folder, typically:
```
/share/piper/
```
2. Copy **both** files for each voice you want into that directory:
```
en_us-cathy-medium.onnx
en_us-cathy-medium.onnx.json
```
3. Restart the Piper add-on. Then in Home Assistant go to
**Settings → Voice assistants → (your assistant) → Text-to-speech**
and select the new voice.
> Tip: If a voice does not appear, confirm the `.onnx` and `.onnx.json`
> filenames match exactly and that both files are present.
## Testing a voice locally
With Piper installed you can render a sample to a WAV file:
```bash
echo "Hello from Home Assistant." | \
piper --model en_us-cathy-medium.onnx --output_file sample.wav
```
## License
The licensing of a trained Piper voice depends on the dataset it was trained
on. Add a `LICENSE` file describing the terms that apply to these models and
their training data before distributing them.
en_us-cathy-medium.onnx
BIN
View File
Binary file not shown.
en_us-cathy-medium.onnx.json
+497
View File
@@ -0,0 +1,497 @@
{
"dataset": "en_us-cathy-medium",
"audio": {
"sample_rate": 22050,
"quality": "training"
},
"espeak": {
"voice": "en-us"
},
"language": {
"code": "en-us"
},
"inference": {
"noise_scale": 0.667,
"length_scale": 1,
"noise_w": 0.8
},
"phoneme_type": "espeak",
"phoneme_map": {},
"phoneme_id_map": {
" ": [
3
],
"!": [
4
],
"\"": [
150
],
"#": [
149
],
"$": [
2
],
"'": [
5
],
"(": [
6
],
")": [
7
],
",": [
8
],
"-": [
9
],
".": [
10
],
"0": [
130
],
"1": [
131
],
"2": [
132
],
"3": [
133
],
"4": [
134
],
"5": [
135
],
"6": [
136
],
"7": [
137
],
"8": [
138
],
"9": [
139
],
":": [
11
],
";": [
12
],
"?": [
13
],
"X": [
156
],
"^": [
1
],
"_": [
0
],
"a": [
14
],
"b": [
15
],
"c": [
16
],
"d": [
17
],
"e": [
18
],
"f": [
19
],
"g": [
154
],
"h": [
20
],
"i": [
21
],
"j": [
22
],
"k": [
23
],
"l": [
24
],
"m": [
25
],
"n": [
26
],
"o": [
27
],
"p": [
28
],
"q": [
29
],
"r": [
30
],
"s": [
31
],
"t": [
32
],
"u": [
33
],
"v": [
34
],
"w": [
35
],
"x": [
36
],
"y": [
37
],
"z": [
38
],
"\u00e6": [
39
],
"\u00e7": [
40
],
"\u00f0": [
41
],
"\u00f8": [
42
],
"\u0127": [
43
],
"\u014b": [
44
],
"\u0153": [
45
],
"\u01c0": [
46
],
"\u01c1": [
47
],
"\u01c2": [
48
],
"\u01c3": [
49
],
"\u0250": [
50
],
"\u0251": [
51
],
"\u0252": [
52
],
"\u0253": [
53
],
"\u0254": [
54
],
"\u0255": [
55
],
"\u0256": [
56
],
"\u0257": [
57
],
"\u0258": [
58
],
"\u0259": [
59
],
"\u025a": [
60
],
"\u025b": [
61
],
"\u025c": [
62
],
"\u025e": [
63
],
"\u025f": [
64
],
"\u0260": [
65
],
"\u0261": [
66
],
"\u0262": [
67
],
"\u0263": [
68
],
"\u0264": [
69
],
"\u0265": [
70
],
"\u0266": [
71
],
"\u0267": [
72
],
"\u0268": [
73
],
"\u026a": [
74
],
"\u026b": [
75
],
"\u026c": [
76
],
"\u026d": [
77
],
"\u026e": [
78
],
"\u026f": [
79
],
"\u0270": [
80
],
"\u0271": [
81
],
"\u0272": [
82
],
"\u0273": [
83
],
"\u0274": [
84
],
"\u0275": [
85
],
"\u0276": [
86
],
"\u0278": [
87
],
"\u0279": [
88
],
"\u027a": [
89
],
"\u027b": [
90
],
"\u027d": [
91
],
"\u027e": [
92
],
"\u0280": [
93
],
"\u0281": [
94
],
"\u0282": [
95
],
"\u0283": [
96
],
"\u0284": [
97
],
"\u0288": [
98
],
"\u0289": [
99
],
"\u028a": [
100
],
"\u028b": [
101
],
"\u028c": [
102
],
"\u028d": [
103
],
"\u028e": [
104
],
"\u028f": [
105
],
"\u0290": [
106
],
"\u0291": [
107
],
"\u0292": [
108
],
"\u0294": [
109
],
"\u0295": [
110
],
"\u0298": [
111
],
"\u0299": [
112
],
"\u029b": [
113
],
"\u029c": [
114
],
"\u029d": [
115
],
"\u029f": [
116
],
"\u02a1": [
117
],
"\u02a2": [
118
],
"\u02a6": [
155
],
"\u02b0": [
145
],
"\u02b2": [
119
],
"\u02c8": [
120
],
"\u02cc": [
121
],
"\u02d0": [
122
],
"\u02d1": [
123
],
"\u02de": [
124
],
"\u02e4": [
146
],
"\u0303": [
141
],
"\u0327": [
140
],
"\u0329": [
144
],
"\u032a": [
142
],
"\u032f": [
143
],
"\u033a": [
152
],
"\u033b": [
153
],
"\u03b2": [
125
],
"\u03b5": [
147
],
"\u03b8": [
126
],
"\u03c7": [
127
],
"\u1d7b": [
128
],
"\u2191": [
151
],
"\u2193": [
148
],
"\u2c71": [
129
]
},
"num_symbols": 256,
"num_speakers": 1,
"speaker_id_map": {},
"piper_version": "1.0.0"
}
en_us-cynda-medium.onnx
BIN
View File
Binary file not shown.
en_us-cynda-medium.onnx.json
+497
View File
@@ -0,0 +1,497 @@
{
"dataset": "en_us-cynda-medium",
"audio": {
"sample_rate": 22050,
"quality": "training"
},
"espeak": {
"voice": "en-us"
},
"language": {
"code": "en-us"
},
"inference": {
"noise_scale": 0.667,
"length_scale": 1,
"noise_w": 0.8
},
"phoneme_type": "espeak",
"phoneme_map": {},
"phoneme_id_map": {
" ": [
3
],
"!": [
4
],
"\"": [
150
],
"#": [
149
],
"$": [
2
],
"'": [
5
],
"(": [
6
],
")": [
7
],
",": [
8
],
"-": [
9
],
".": [
10
],
"0": [
130
],
"1": [
131
],
"2": [
132
],
"3": [
133
],
"4": [
134
],
"5": [
135
],
"6": [
136
],
"7": [
137
],
"8": [
138
],
"9": [
139
],
":": [
11
],
";": [
12
],
"?": [
13
],
"X": [
156
],
"^": [
1
],
"_": [
0
],
"a": [
14
],
"b": [
15
],
"c": [
16
],
"d": [
17
],
"e": [
18
],
"f": [
19
],
"g": [
154
],
"h": [
20
],
"i": [
21
],
"j": [
22
],
"k": [
23
],
"l": [
24
],
"m": [
25
],
"n": [
26
],
"o": [
27
],
"p": [
28
],
"q": [
29
],
"r": [
30
],
"s": [
31
],
"t": [
32
],
"u": [
33
],
"v": [
34
],
"w": [
35
],
"x": [
36
],
"y": [
37
],
"z": [
38
],
"\u00e6": [
39
],
"\u00e7": [
40
],
"\u00f0": [
41
],
"\u00f8": [
42
],
"\u0127": [
43
],
"\u014b": [
44
],
"\u0153": [
45
],
"\u01c0": [
46
],
"\u01c1": [
47
],
"\u01c2": [
48
],
"\u01c3": [
49
],
"\u0250": [
50
],
"\u0251": [
51
],
"\u0252": [
52
],
"\u0253": [
53
],
"\u0254": [
54
],
"\u0255": [
55
],
"\u0256": [
56
],
"\u0257": [
57
],
"\u0258": [
58
],
"\u0259": [
59
],
"\u025a": [
60
],
"\u025b": [
61
],
"\u025c": [
62
],
"\u025e": [
63
],
"\u025f": [
64
],
"\u0260": [
65
],
"\u0261": [
66
],
"\u0262": [
67
],
"\u0263": [
68
],
"\u0264": [
69
],
"\u0265": [
70
],
"\u0266": [
71
],
"\u0267": [
72
],
"\u0268": [
73
],
"\u026a": [
74
],
"\u026b": [
75
],
"\u026c": [
76
],
"\u026d": [
77
],
"\u026e": [
78
],
"\u026f": [
79
],
"\u0270": [
80
],
"\u0271": [
81
],
"\u0272": [
82
],
"\u0273": [
83
],
"\u0274": [
84
],
"\u0275": [
85
],
"\u0276": [
86
],
"\u0278": [
87
],
"\u0279": [
88
],
"\u027a": [
89
],
"\u027b": [
90
],
"\u027d": [
91
],
"\u027e": [
92
],
"\u0280": [
93
],
"\u0281": [
94
],
"\u0282": [
95
],
"\u0283": [
96
],
"\u0284": [
97
],
"\u0288": [
98
],
"\u0289": [
99
],
"\u028a": [
100
],
"\u028b": [
101
],
"\u028c": [
102
],
"\u028d": [
103
],
"\u028e": [
104
],
"\u028f": [
105
],
"\u0290": [
106
],
"\u0291": [
107
],
"\u0292": [
108
],
"\u0294": [
109
],
"\u0295": [
110
],
"\u0298": [
111
],
"\u0299": [
112
],
"\u029b": [
113
],
"\u029c": [
114
],
"\u029d": [
115
],
"\u029f": [
116
],
"\u02a1": [
117
],
"\u02a2": [
118
],
"\u02a6": [
155
],
"\u02b0": [
145
],
"\u02b2": [
119
],
"\u02c8": [
120
],
"\u02cc": [
121
],
"\u02d0": [
122
],
"\u02d1": [
123
],
"\u02de": [
124
],
"\u02e4": [
146
],
"\u0303": [
141
],
"\u0327": [
140
],
"\u0329": [
144
],
"\u032a": [
142
],
"\u032f": [
143
],
"\u033a": [
152
],
"\u033b": [
153
],
"\u03b2": [
125
],
"\u03b5": [
147
],
"\u03b8": [
126
],
"\u03c7": [
127
],
"\u1d7b": [
128
],
"\u2191": [
151
],
"\u2193": [
148
],
"\u2c71": [
129
]
},
"num_symbols": 256,
"num_speakers": 1,
"speaker_id_map": {},
"piper_version": "1.0.0"
}
en_us-james-medium.onnx
BIN
View File
Binary file not shown.
en_us-james-medium.onnx.json
+497
View File
@@ -0,0 +1,497 @@
{
"dataset": "en_us-james-medium",
"audio": {
"sample_rate": 22050,
"quality": "training"
},
"espeak": {
"voice": "en-us"
},
"language": {
"code": "en-us"
},
"inference": {
"noise_scale": 0.667,
"length_scale": 1,
"noise_w": 0.8
},
"phoneme_type": "espeak",
"phoneme_map": {},
"phoneme_id_map": {
" ": [
3
],
"!": [
4
],
"\"": [
150
],
"#": [
149
],
"$": [
2
],
"'": [
5
],
"(": [
6
],
")": [
7
],
",": [
8
],
"-": [
9
],
".": [
10
],
"0": [
130
],
"1": [
131
],
"2": [
132
],
"3": [
133
],
"4": [
134
],
"5": [
135
],
"6": [
136
],
"7": [
137
],
"8": [
138
],
"9": [
139
],
":": [
11
],
";": [
12
],
"?": [
13
],
"X": [
156
],
"^": [
1
],
"_": [
0
],
"a": [
14
],
"b": [
15
],
"c": [
16
],
"d": [
17
],
"e": [
18
],
"f": [
19
],
"g": [
154
],
"h": [
20
],
"i": [
21
],
"j": [
22
],
"k": [
23
],
"l": [
24
],
"m": [
25
],
"n": [
26
],
"o": [
27
],
"p": [
28
],
"q": [
29
],
"r": [
30
],
"s": [
31
],
"t": [
32
],
"u": [
33
],
"v": [
34
],
"w": [
35
],
"x": [
36
],
"y": [
37
],
"z": [
38
],
"\u00e6": [
39
],
"\u00e7": [
40
],
"\u00f0": [
41
],
"\u00f8": [
42
],
"\u0127": [
43
],
"\u014b": [
44
],
"\u0153": [
45
],
"\u01c0": [
46
],
"\u01c1": [
47
],
"\u01c2": [
48
],
"\u01c3": [
49
],
"\u0250": [
50
],
"\u0251": [
51
],
"\u0252": [
52
],
"\u0253": [
53
],
"\u0254": [
54
],
"\u0255": [
55
],
"\u0256": [
56
],
"\u0257": [
57
],
"\u0258": [
58
],
"\u0259": [
59
],
"\u025a": [
60
],
"\u025b": [
61
],
"\u025c": [
62
],
"\u025e": [
63
],
"\u025f": [
64
],
"\u0260": [
65
],
"\u0261": [
66
],
"\u0262": [
67
],
"\u0263": [
68
],
"\u0264": [
69
],
"\u0265": [
70
],
"\u0266": [
71
],
"\u0267": [
72
],
"\u0268": [
73
],
"\u026a": [
74
],
"\u026b": [
75
],
"\u026c": [
76
],
"\u026d": [
77
],
"\u026e": [
78
],
"\u026f": [
79
],
"\u0270": [
80
],
"\u0271": [
81
],
"\u0272": [
82
],
"\u0273": [
83
],
"\u0274": [
84
],
"\u0275": [
85
],
"\u0276": [
86
],
"\u0278": [
87
],
"\u0279": [
88
],
"\u027a": [
89
],
"\u027b": [
90
],
"\u027d": [
91
],
"\u027e": [
92
],
"\u0280": [
93
],
"\u0281": [
94
],
"\u0282": [
95
],
"\u0283": [
96
],
"\u0284": [
97
],
"\u0288": [
98
],
"\u0289": [
99
],
"\u028a": [
100
],
"\u028b": [
101
],
"\u028c": [
102
],
"\u028d": [
103
],
"\u028e": [
104
],
"\u028f": [
105
],
"\u0290": [
106
],
"\u0291": [
107
],
"\u0292": [
108
],
"\u0294": [
109
],
"\u0295": [
110
],
"\u0298": [
111
],
"\u0299": [
112
],
"\u029b": [
113
],
"\u029c": [
114
],
"\u029d": [
115
],
"\u029f": [
116
],
"\u02a1": [
117
],
"\u02a2": [
118
],
"\u02a6": [
155
],
"\u02b0": [
145
],
"\u02b2": [
119
],
"\u02c8": [
120
],
"\u02cc": [
121
],
"\u02d0": [
122
],
"\u02d1": [
123
],
"\u02de": [
124
],
"\u02e4": [
146
],
"\u0303": [
141
],
"\u0327": [
140
],
"\u0329": [
144
],
"\u032a": [
142
],
"\u032f": [
143
],
"\u033a": [
152
],
"\u033b": [
153
],
"\u03b2": [
125
],
"\u03b5": [
147
],
"\u03b8": [
126
],
"\u03c7": [
127
],
"\u1d7b": [
128
],
"\u2191": [
151
],
"\u2193": [
148
],
"\u2c71": [
129
]
},
"num_symbols": 256,
"num_speakers": 1,
"speaker_id_map": {},
"piper_version": "1.0.0"
}