Enhancing the Stability of LLM-based Speech Generation Systems through Self-Supervised Representations
By Álvaro Martín-Cortinas, Daniel Sáez-Trigueros, Iv'an Vall'es-P'erez, Biel Tura Vecino, Piotr Bilinski, Mateusz Lajszczak, Grzegorz Beringer, Roberto Barra-Chicote, Jaime Lorenzo-Trueba. 2024
9 citations recorded by the source corpus.
Large Language Models (LLMs) are one of the most promising technologies for the next era of speech generation systems, due to their scalability and in-context learning capabilities. Nevertheless, they suffer from multiple stability issues at inference time, such as hallucinations, content skipping or speech repetitions. In this work, we introduce a new self-supervised Voice Conversion (VC) architecture which can be used to learn to encode transitory features, such as content, separately from stationary ones, such as speaker ID or recording conditions, creating speaker-disentangled representations. Using speaker-disentangled codes to train LLMs for text-to-speech (TTS) allows the LLM to generate the content and the style of the speech only from the text, similarly to humans, while the speaker identity is provided by the decoder of the VC model. Results show that LLMs trained over speaker-disentangled self-supervised representations provide an improvement of 4.7pp in speaker similarity over SOTA entangled representations, and a word error rate (WER) 5.4pp lower. Furthermore, they achieve higher naturalness than human recordings of the LibriTTS test-other dataset. Finally, we show that using explicit reference embedding negatively impacts intelligibility (stability), with WER increasing by 14pp compared to the model that only uses text to infer the style.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sri Karlapati
- Alexis Moinet
- Thomas Drugman
- Bartosz Putrycz
- Ewa Muszynska
- Mateusz Lajszczak
- Zhenhua Ling
- Yang Ai
Citations in the corpus
- BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of data (2024, methodology): “Our approach similar to the one introduced in [43] with modifications that reduce bitrate of the codes.”