Puyuan Peng

Zero-Shot Speech Editing and Text-to-Speech in the Wild

VoiceCraft is an AI model that clones a person's voice from a short audio sample, then generates new spoken words or edits existing recordings to sound like that person, working on real-world audio like podcasts and audiobooks.

VoiceStar: Robust, Duration-controllable TTS that can Extrapolate

Promting Whisper for Audio-Visual Speech Recognition, Code-Switched Speech Recognition, and Zero-Shot Speech Translation

Syllable Segmentation and Cross-Lingual Generalization in a Visually Grounded, Self-Supervised Speech Model

Word Discovery in Visually Grounded, Self-Supervised Speech Models

Transformer-based visually grounded speech models

No description.

No description.

[NeurIPS 2021] Moment-DETR code and QVHighlights dataset

No description.

Github Pages template for academic personal websites, forked from mmistakes/minimal-mistakes

No description.

Public Code for the paper MAE-AST: Masked Autoencoding Audio Spectrogram Transformer

Unsupervised phone and word segmentation using dynamic programming on self-supervised VQ features.

Video Feature Extraction Code for EMNLP 2020 paper "HERO: Hierarchical Encoder for Video+Language Omni-representation Pre-training"

No description.

BERT and LSTM baseline models of the ZeroSpeech Challenge 2021

Pre-trained models and code and data to train and use models from "Pushing the Limits of Paraphrastic Sentence Embeddings with Millions of Machine Translations"