Spirit LM: Interleaved Spoken and Written Language Model

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Nguyen, Tu Anh, Muller, Benjamin, Yu, Bokai, Costa-jussa, Marta R., Elbayad, Maha, Popuri, Sravya, Ropers, Christophe, Duquenne, Paul-Ambroise, Algayres, Robin, Mavlyutov, Ruslan, Gat, Itai, Williamson, Mary, Synnaeve, Gabriel, Pino, Juan, Sagot, Benoit, Dupoux, Emmanuel
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916445249601536
author Nguyen, Tu Anh
Muller, Benjamin
Yu, Bokai
Costa-jussa, Marta R.
Elbayad, Maha
Popuri, Sravya
Ropers, Christophe
Duquenne, Paul-Ambroise
Algayres, Robin
Mavlyutov, Ruslan
Gat, Itai
Williamson, Mary
Synnaeve, Gabriel
Pino, Juan
Sagot, Benoit
Dupoux, Emmanuel
author_facet Nguyen, Tu Anh
Muller, Benjamin
Yu, Bokai
Costa-jussa, Marta R.
Elbayad, Maha
Popuri, Sravya
Ropers, Christophe
Duquenne, Paul-Ambroise
Algayres, Robin
Mavlyutov, Ruslan
Gat, Itai
Williamson, Mary
Synnaeve, Gabriel
Pino, Juan
Sagot, Benoit
Dupoux, Emmanuel
contents We introduce Spirit LM, a foundation multimodal language model that freely mixes text and speech. Our model is based on a 7B pretrained text language model that we extend to the speech modality by continuously training it on text and speech units. Speech and text sequences are concatenated as a single stream of tokens, and trained with a word-level interleaving method using a small automatically-curated speech-text parallel corpus. Spirit LM comes in two versions: a Base version that uses speech phonetic units (HuBERT) and an Expressive version that models expressivity using pitch and style units in addition to the phonetic units. For both versions, the text is encoded with subword BPE tokens. The resulting model displays both the semantic abilities of text models and the expressive abilities of speech models. Additionally, we demonstrate that Spirit LM can learn new tasks in a few-shot fashion across modalities (i.e. ASR, TTS, Speech Classification). We make available model weights and inference code.
format Preprint
id arxiv_https___arxiv_org_abs_2402_05755
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Spirit LM: Interleaved Spoken and Written Language Model
Nguyen, Tu Anh
Muller, Benjamin
Yu, Bokai
Costa-jussa, Marta R.
Elbayad, Maha
Popuri, Sravya
Ropers, Christophe
Duquenne, Paul-Ambroise
Algayres, Robin
Mavlyutov, Ruslan
Gat, Itai
Williamson, Mary
Synnaeve, Gabriel
Pino, Juan
Sagot, Benoit
Dupoux, Emmanuel
Computation and Language
Sound
Audio and Speech Processing
We introduce Spirit LM, a foundation multimodal language model that freely mixes text and speech. Our model is based on a 7B pretrained text language model that we extend to the speech modality by continuously training it on text and speech units. Speech and text sequences are concatenated as a single stream of tokens, and trained with a word-level interleaving method using a small automatically-curated speech-text parallel corpus. Spirit LM comes in two versions: a Base version that uses speech phonetic units (HuBERT) and an Expressive version that models expressivity using pitch and style units in addition to the phonetic units. For both versions, the text is encoded with subword BPE tokens. The resulting model displays both the semantic abilities of text models and the expressive abilities of speech models. Additionally, we demonstrate that Spirit LM can learn new tasks in a few-shot fashion across modalities (i.e. ASR, TTS, Speech Classification). We make available model weights and inference code.
title Spirit LM: Interleaved Spoken and Written Language Model
topic Computation and Language
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2402.05755