Although the recent development of LLMs has paved the way for multi-modal interaction, minority languages face difficulties keeping pace. This work presents the first open Speech Language Model for Basque, designed for low-latency, end-to-end spoken interaction. Based on the LLaMA-Omni architecture, the system integrates a Latxa model as its backbone, transferring its Basque capabilities to the spoken modality by fine-tuning several modules. To address data scarcity, the English VoiceAssistant-400K corpus was translated into Basque, and the audio was synthesized using a Basque TTS. Evaluation results confirm the system's viability, achieving 3.86/5 in content quality, a 9.94% WER for spoken intelligibility, and a 3.07 UTMOS for audio quality. This work demonstrates that advanced spoken interaction systems can be developed using modular architectures and synthetic data even in low-resource languages.
Figures & tables
Figure 1: Left : Model architecture of our SLM, based on LLaMA-Omni. Right : Illustration of the two-stage training strategy.
Model
LLM
WER (%) ↓
UTMOS ↑
Streaming
Score (1-5) ↑
Latency (ms) ↓
SpeechGPT (Eng)
2.98
45.00
3.89
5587
SALMONN + Orca (Eng)
3.44
3.78
3.82
-
Qwen2-Audio + Orca (Eng)
3.47
6.77
3.62
-
LLaMA-Omni (Eng)
3.99
10.82
3.92
346
LLaMA-Omni2-7B (Eng)
4.28
3.26
4.22
582
Table 1: Comparison of our Basque (Eus) model with English (Eng) SLMs. English values are as reported in [ 1 ] and [ 2 ] , using different judges/ASR systems; WER on synthesized speech also depends on TTS and ASR quality, so cross-language figures are indicative, not strictly comparable.