Uniform steps are definitely a problem in speech LLMs, couple of attempts to solve that which come together recently, the idea is that we apply text/speech alignment before we feed data into LLM:

https://github.com/FreedomIntelligence/Soundwave

https://github.com/mtkresearch/TASTE-SpokenLM

https://arxiv.org/abs/2502.12900

Soundwave: Less is More for Speech-Text Alignment in LLMs

Yuhao Zhang, Zhiheng Liu, Fan Bu, Ruiyu Zhang, Benyou Wang, Haizhou Li

Existing end-to-end speech large language models (LLMs) usually rely on large-scale annotated data for training, while data-efficient training has not been discussed in depth. We focus on two fundamental problems between speech and text: the representation space gap and sequence length inconsistency. We propose Soundwave, which utilizes an efficient training strategy and a novel architecture to address these issues. Results show that Soundwave outperforms the advanced Qwen2-Audio in speech translation and AIR-Bench speech tasks, using only one-fiftieth of the training data. Further analysis shows that Soundwave still retains its intelligence during conversation. The project is available at this https URL.

Speech Technology

Uniform steps are definitely a problem in speech LLMs, couple of attempts to solve that which come together recently, the idea is that we apply text/speech alignment before we feed data into LLM:

https://github.com/FreedomIntelligence/Soundwave

https://github.com/mtkresearch/TASTE-SpokenLM

https://arxiv.org/abs/2502.12900

Soundwave: Less is More for Speech-Text Alignment in LLMs

Yuhao Zhang, Zhiheng Liu, Fan Bu, Ruiyu Zhang, Benyou Wang, Haizhou Li

Existing end-to-end speech large language models (LLMs) usually rely on large-scale annotated data for training, while data-efficient training has not been discussed in depth. We focus on two fundamental problems between speech and text: the representation space gap and sequence length inconsistency. We propose Soundwave, which utilizes an efficient training strategy and a novel architecture to address these issues. Results show that Soundwave outperforms the advanced Qwen2-Audio in speech translation and AIR-Bench speech tasks, using only one-fiftieth of the training data. Further analysis shows that Soundwave still retains its intelligence during conversation. The project is available at this https URL.

GitHub

GitHub - FreedomIntelligence/Soundwave: The official Soundwave repository

The official Soundwave repository. Contribute to FreedomIntelligence/Soundwave development by creating an account on GitHub.

hottg.com/speechtech/2099

1.9K viewsApr 11 at 21:58