配音致谢
ClipKart 使用免费的开放配音。以下是它使用的引擎和声音、训练所用的数据以及各自的许可证。
引擎
- Kokoro-82M · Apache-2.0
- ElevenLabs Flash v2.5 · ElevenLabs terms, paid plan
- Piper · GPL-3.0 engine, run as a separate process and never distributed; each voice has its own licence
声音
| 声音 | 语言 | 训练数据 | 许可证 |
|---|---|---|---|
| Heart af_heart · Kokoro | English (en-US) | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Michael am_michael · Kokoro | English (en-US) | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Emma bf_emma · Kokoro | English (en-GB) | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Sara if_sara · Kokoro | Italiano | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Nicola im_nicola · Kokoro | Italiano | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Dora ef_dora · Kokoro | Español | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Alex em_alex · Kokoro | Español | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Dora pf_dora · Kokoro | Português (Portugal) (pt-BR) | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Alex pm_alex · Kokoro | Português (Portugal) (pt-BR) | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Xiaobei zf_xiaobei · Kokoro | 简体中文 | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| Yunjian zm_yunjian · Kokoro | 简体中文 | Kokoro-82M by hexgrad, trained on permissive or non-copyrighted audio (model card) | Apache-2.0 |
| MLS de_DE-mls-medium · Piper | Deutsch | Multilingual LibriSpeech (Pratap et al., 2020), OpenSLR 94; trained from scratch | CC BY 4.0 |
| MLS nl_NL-mls-medium · Piper | Nederlands | Multilingual LibriSpeech (Pratap et al., 2020), OpenSLR 94; trained from scratch | CC BY 4.0 |
| NST sv_SE-nst-medium · Piper | Svenska | NST Swedish speech synthesis, Språkbanken (National Library of Norway); trained from scratch by KBLab | CC0 |
以下语言暂不提供配音：Polski, Dansk, Čeština, বাংলা。