Text-Guided HuBERT: Self-Supervised Speech Pre-Training via Generative Adversarial Networks.
Duo MaXianghu YueJunyi AoXiaoxue GaoHaizhou LiPublished in: IEEE Signal Process. Lett. (2024)
Keyphrases
- text to speech
- text to speech synthesis
- english text
- speech recognition
- multi lingual
- hearing impaired
- training process
- discriminative training
- text classifiers
- text retrieval
- social networks
- text input
- echo state networks
- training corpus
- automatic speech recognition
- free text
- database
- training set
- web documents
- complex networks
- network structure
- spoken documents
- text mining
- spontaneous speech
- text recognition
- speech synthesis
- recurrent networks
- text documents
- unsupervised learning
- training examples
- speech signal
- generative model
- lexical features
- spoken language
- community structure
- audio visual
- text data