| license: apache-2.0 | |
| The _tortoise-tts-ruslan_ model is a tortoise model capable of speaking Russian language. | |
| The model was first finetuned on public_youtube700_val+buriy_audiobooks_2_val from [Russian Open Speech To Text](https://learn.microsoft.com/en-us/azure/open-datasets/dataset-open-speech-text?tabs=azure-storage). | |
| The model was then finetuned on [RUSLAN: Russian Spoken Language Corpus For Speech Synthesis](https://ruslan-corpus.github.io/). | |
| The model is able to generate generic male voices, see [examples](https://huggingface.co/SerCe/tortoise-tts-ruslan/tree/main/examples/random). | |
| Additionally, the model is suitable for further finetuning on any Russian male voice, e.g. see a [finetuned](https://huggingface.co/SerCe/tortoise-tts-ruslan/tree/main/examples/finetuned) voice of Yury Dud (the finetuned model weights are not included). | |
| The code from [ai-voice-cloning](https://git.ecker.tech/mrq/ai-voice-cloning) was used to train the model. |