Multi-Tacotron-Voice-Cloning and Cross-Lingual-Voice-Cloning
About Multi-Tacotron-Voice-Cloning
vlomme/Multi-Tacotron-Voice-Cloning
Phoneme multilingual(Russian-English) voice cloning based on
Implements a four-stage pipeline combining speaker verification (GE2E encoder), phonemic text-to-speech synthesis (Tacotron 2), and neural vocoding (WaveRNN) to enable few-shot voice cloning from seconds of audio. Uses phoneme-based representation with language-specific dictionaries to support both Russian and English in a unified model. Provides pretrained weights and curated multilingual datasets, with training extensible to additional languages via the phoneme dictionary approach.
About Cross-Lingual-Voice-Cloning
deterministic-algorithms-lab/Cross-Lingual-Voice-Cloning
Tacotron 2 - PyTorch implementation with faster-than-realtime inference modified to enable cross lingual voice cloning.
This tool helps content creators and educators generate speech in a new language using an existing speaker's voice. You provide audio files of a person speaking along with corresponding text, and the system learns to clone their voice. The output is a synthesized audio file of that speaker's voice speaking a different language from the input, making it useful for creating consistent voiceovers or narrated content across multiple languages.
Scores updated daily from GitHub, PyPI, and npm data. How scores work