| 1 | # Copyright (C) 2025 AIDC-AI |
| 2 | # |
| 3 | # Licensed under the Apache License, Version 2.0 (the "License"); |
| 4 | # you may not use this file except in compliance with the License. |
| 5 | # You may obtain a copy of the License at |
| 6 | # http://www.apache.org/licenses/LICENSE-2.0 |
| 7 | # Unless required by applicable law or agreed to in writing, software |
| 8 | # distributed under the License is distributed on an "AS IS" BASIS, |
| 9 | # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 10 | # See the License for the specific language governing permissions and |
| 11 | # limitations under the License. |
| 12 | |
| 13 | """ |
| 14 | TTS Voice Configuration |
| 15 | |
| 16 | Defines available voices for local Edge TTS inference. |
| 17 | """ |
| 18 | |
| 19 | from typing import List, Dict, Any |
| 20 | |
| 21 | |
| 22 | # Edge TTS voice presets for local inference |
| 23 | EDGE_TTS_VOICES: List[Dict[str, Any]] = [ |
| 24 | # Chinese voices |
| 25 | { |
| 26 | "id": "zh-CN-XiaoxiaoNeural", |
| 27 | "label_key": "tts.voice.zh_CN_XiaoxiaoNeural", |
| 28 | "locale": "zh-CN", |
| 29 | "gender": "female" |
| 30 | }, |
| 31 | { |
| 32 | "id": "zh-CN-XiaoyiNeural", |
| 33 | "label_key": "tts.voice.zh_CN_XiaoyiNeural", |
| 34 | "locale": "zh-CN", |
| 35 | "gender": "female" |
| 36 | }, |
| 37 | { |
| 38 | "id": "zh-CN-YunjianNeural", |
| 39 | "label_key": "tts.voice.zh_CN_YunjianNeural", |
| 40 | "locale": "zh-CN", |
| 41 | "gender": "male" |
| 42 | }, |
| 43 | { |
| 44 | "id": "zh-CN-YunxiNeural", |
| 45 | "label_key": "tts.voice.zh_CN_YunxiNeural", |
| 46 | "locale": "zh-CN", |
| 47 | "gender": "male" |
| 48 | }, |
| 49 | { |
| 50 | "id": "zh-CN-YunyangNeural", |
| 51 | "label_key": "tts.voice.zh_CN_YunyangNeural", |
| 52 | "locale": "zh-CN", |
| 53 | "gender": "male" |
| 54 | }, |
| 55 | { |
| 56 | "id": "zh-CN-YunyeNeural", |
| 57 | "label_key": "tts.voice.zh_CN_YunyeNeural", |
| 58 | "locale": "zh-CN", |
| 59 | "gender": "male" |
| 60 | }, |
| 61 | { |
| 62 | "id": "zh-CN-YunfengNeural", |
| 63 | "label_key": "tts.voice.zh_CN_YunfengNeural", |
| 64 | "locale": "zh-CN", |
| 65 | "gender": "male" |
| 66 | }, |
| 67 | { |
| 68 | "id": "zh-CN-liaoning-XiaobeiNeural", |
| 69 | "label_key": "tts.voice.zh_CN_liaoning_XiaobeiNeural", |
| 70 | "locale": "zh-CN", |
| 71 | "gender": "female" |
| 72 | }, |
| 73 | { |
| 74 | "id": "en-US-AriaNeural", |
| 75 | "label_key": "tts.voice.en_US_AriaNeural", |
| 76 | "locale": "en-US", |
| 77 | "gender": "female" |
| 78 | }, |
| 79 | { |
| 80 | "id": "en-US-JennyNeural", |
| 81 | "label_key": "tts.voice.en_US_JennyNeural", |
| 82 | "locale": "en-US", |
| 83 | "gender": "female" |
| 84 | }, |
| 85 | { |
| 86 | "id": "en-US-GuyNeural", |
| 87 | "label_key": "tts.voice.en_US_GuyNeural", |
| 88 | "locale": "en-US", |
| 89 | "gender": "male" |
| 90 | }, |
| 91 | { |
| 92 | "id": "en-US-DavisNeural", |
| 93 | "label_key": "tts.voice.en_US_DavisNeural", |
| 94 | "locale": "en-US", |
| 95 | "gender": "male" |
| 96 | }, |
| 97 | { |
| 98 | "id": "en-GB-SoniaNeural", |
| 99 | "label_key": "tts.voice.en_GB_SoniaNeural", |
| 100 | "locale": "en-GB", |
| 101 | "gender": "female" |
| 102 | }, |
| 103 | { |
| 104 | "id": "en-GB-RyanNeural", |
| 105 | "label_key": "tts.voice.en_GB_RyanNeural", |
| 106 | "locale": "en-GB", |
| 107 | "gender": "male" |
| 108 | }, |
| 109 | { |
| 110 | "id": "ko-KR-InJoonNeural", |
| 111 | "label_key": "tts.voice.ko-KR-InJoonNeural", |
| 112 | "locale": "ko-KR", |
| 113 | "gender": "male" |
| 114 | }, |
| 115 | { |
| 116 | "id": "ko-KR-SunHiNeural", |
| 117 | "label_key": "tts.voice.ko-KR-SunHiNeural", |
| 118 | "locale": "ko-KR", |
| 119 | "gender": "female" |
| 120 | }, |
| 121 | { |
| 122 | "id": "fr-FR-EloiseNeural", |
| 123 | "label_key": "tts.voice.fr-FR-EloiseNeural", |
| 124 | "locale": "fr-FR", |
| 125 | "gender": "female" |
| 126 | }, |
| 127 | { |
| 128 | "id": "fr-FR-HenriNeural", |
| 129 | "label_key": "tts.voice.fr-FR-HenriNeural", |
| 130 | "locale": "fr-FR", |
| 131 | "gender": "male" |
| 132 | }, |
| 133 | { |
| 134 | "id": "pt-PT-DuarteNeural", |
| 135 | "label_key": "tts.voice.pt-PT-DuarteNeural", |
| 136 | "locale": "pt-PT", |
| 137 | "gender": "male" |
| 138 | }, |
| 139 | { |
| 140 | "id": "pt-PT-RaquelNeural", |
| 141 | "label_key": "tts.voice.pt-PT-RaquelNeural", |
| 142 | "locale": "pt-PT", |
| 143 | "gender": "female" |
| 144 | }, |
| 145 | { |
| 146 | "id": "de-DE-AmalaNeural", |
| 147 | "label_key": "tts.voice.de-DE-AmalaNeural", |
| 148 | "locale": "de-DE", |
| 149 | "gender": "female" |
| 150 | }, |
| 151 | { |
| 152 | "id": "de-DE-ConradNeural", |
| 153 | "label_key": "tts.voice.de-DE-ConradNeural", |
| 154 | "locale": "de-DE", |
| 155 | "gender": "male" |
| 156 | }, |
| 157 | |
| 158 | # English voices |
| 159 | { |
| 160 | "id": "ru-RU-DmitryNeural", |
| 161 | "label_key": "tts.voice.ru-RU-DmitryNeural", |
| 162 | "locale": "ru-RU", |
| 163 | "gender": "male" |
| 164 | }, |
| 165 | { |
| 166 | "id": "ru-RU-SvetlanaNeural", |
| 167 | "label_key": "tts.voice.ru-RU-SvetlanaNeural", |
| 168 | "locale": "ru-RU", |
| 169 | "gender": "female" |
| 170 | }, |
| 171 | { |
| 172 | "id": "tr-TR-AhmetNeural", |
| 173 | "label_key": "tts.voice.tr-TR-AhmetNeural", |
| 174 | "locale": "tr-TR", |
| 175 | "gender": "male" |
| 176 | }, |
| 177 | { |
| 178 | "id": "tr-TR-EmelNeural", |
| 179 | "label_key": "tts.voice.tr-TR-EmelNeural", |
| 180 | "locale": "tr-TR", |
| 181 | "gender": "female" |
| 182 | }, |
| 183 | { |
| 184 | "id": "es-ES-AlvaroNeural", |
| 185 | "label_key": "tts.voice.es-ES-AlvaroNeural", |
| 186 | "locale": "es-ES", |
| 187 | "gender": "male" |
| 188 | }, |
| 189 | { |
| 190 | "id": "es-ES-ElviraNeural", |
| 191 | "label_key": "tts.voice.es-ES-ElviraNeural", |
| 192 | "locale": "es-ES", |
| 193 | "gender": "female" |
| 194 | }, |
| 195 | ] |
| 196 | |
| 197 | |
| 198 | def get_voice_display_name(voice_id: str, tr_func=None, locale: str = "zh_CN") -> str: |
| 199 | """ |
| 200 | Get display name for voice |
| 201 | |
| 202 | Args: |
| 203 | voice_id: Voice ID (e.g., "zh-CN-YunjianNeural") |
| 204 | tr_func: Translation function (optional) |
| 205 | locale: Current locale (default: "zh_CN") |
| 206 | |
| 207 | Returns: |
| 208 | Display name (translated label if in Chinese, otherwise voice ID) |
| 209 | """ |
| 210 | # Find voice config |
| 211 | voice_config = next((v for v in EDGE_TTS_VOICES if v["id"] == voice_id), None) |
| 212 | |
| 213 | if not voice_config: |
| 214 | return voice_id |
| 215 | |
| 216 | # If Chinese locale and translation function available, use translated label |
| 217 | if locale == "zh_CN" and tr_func: |
| 218 | label_key = voice_config["label_key"] |
| 219 | return tr_func(label_key) |
| 220 | |
| 221 | # For other locales, return voice ID |
| 222 | return voice_id |
| 223 | |
| 224 | |
| 225 | def speed_to_rate(speed: float) -> str: |
| 226 | """ |
| 227 | Convert speed multiplier to Edge TTS rate parameter |
| 228 | |
| 229 | Args: |
| 230 | speed: Speed multiplier (1.0 = normal, 1.2 = 120%) |
| 231 | |
| 232 | Returns: |
| 233 | Rate string (e.g., "+20%", "-10%") |
| 234 | |
| 235 | Examples: |
| 236 | 1.0 → "+0%" |
| 237 | 1.2 → "+20%" |
| 238 | 0.8 → "-20%" |
| 239 | """ |
| 240 | percentage = int((speed - 1.0) * 100) |
| 241 | sign = "+" if percentage >= 0 else "" |
| 242 | return f"{sign}{percentage}%" |
| 243 | |
| 244 |