Skip to content
SPECIFICTION
Get listed

VoiceCraft-X

voicecraft-x.github.io

Visit voicecraft-x.github.io
Spec score 6.8 #220 of 233 in Text-to-speech tools

Spec sheet

Purpose
VoiceCraft-X combines multilingual speech editing with zero-shot text-to-speech synthesis in one autoregressive neural codec language model.Read 13 Oct 2026 · voicecraft-x.github.io
Product
VoiceCraft-X is an autoregressive neural codec language model for multilingual speech editing and zero-shot text-to-speech synthesis.Read 12 Oct 2026 · voicecraft-x.github.io
Platforms
self_hostedRead 20 Sep 2026 · voicecraft-x.github.io
License
The work is licensed under Creative Commons Attribution-NonCommercial 4.0 International, and the repository says to contact the authors for commercial use.Read 13 Oct 2026 · github.com
Authorship
The repository credits Zhisheng Zheng and the University of Texas at Austin with copyright in 2025.Read 13 Oct 2026 · github.com
Availability
The repository's listed inference examples are speech editing and speech synthesis Jupyter notebooks.Read 13 Oct 2026 · github.com
Checkpoints
The repository links to a default multilingual checkpoint and English and Chinese monolingual checkpoints, with more languages noted as forthcoming.Read 13 Oct 2026 · github.com
Code-switching
The site shows Chinese-English code-switching samples and says Spanish-English and French-English samples are coming soon.Read 13 Oct 2026 · voicecraft-x.github.io
Commercial use
NoRead 20 Sep 2026 · voicecraft-x.github.io
Consent
The repository says speech generation or editing should not be performed using someone's speech without that person's consent.Read 13 Oct 2026 · github.com
Development status
The repository lists environment setup, inference code, additional monolingual checkpoints, hosted demos, Colab notebooks, command-line support, and efficiency improvements as TODO items.Read 13 Oct 2026 · github.com
Fine-tuning
A code-switching sample table includes a VoiceCraft-X (10h fine-tuning) column.Read 12 Oct 2026 · voicecraft-x.github.io
Input and training
The system overview caption says the model takes text and a speaker embedding as input and is trained to predict speech-token sequences.Read 12 Oct 2026 · voicecraft-x.github.io
Languages
The site lists English, Mandarin, Korean, Japanese, Spanish, French, German, Dutch, Italian, Portuguese, and Polish.Read 13 Oct 2026 · voicecraft-x.github.io
Model architecture
The site says VoiceCraft-X uses the Qwen3 large language model for phoneme-free cross-lingual text processing.Read 12 Oct 2026 · voicecraft-x.github.io
Setup
The repository installation instructions create a Python 3.10 Conda environment and install dependencies with pip.Read 13 Oct 2026 · github.com
Speech editing
The site presents examples of editing existing speech by changing or adding words in a transcript.Read 12 Oct 2026 · voicecraft-x.github.io
Speech editing dependency
Montreal Forced Aligner is used for speech editing, and the repository says it can be skipped if installation is difficult.Read 13 Oct 2026 · github.com
Speech generation
Its token reordering mechanism uses time-aligned text and speech tokens to frame editing and synthesis as sequence generation.Read 13 Oct 2026 · voicecraft-x.github.io
Text processing
The model uses Qwen3 for phoneme-free cross-lingual text processing.Read 13 Oct 2026 · voicecraft-x.github.io
Token mechanism
The site describes a token reordering mechanism using time-aligned text and speech tokens.Read 12 Oct 2026 · voicecraft-x.github.io
Voice cloning
The project describes its speech synthesis as voice-cloning and zero-shot TTS.Read 13 Oct 2026 · voicecraft-x.github.io

Where Specifiction places it

Alternatives to VoiceCraft-XAll alternatives →