1 citations · 2 across the 5 of their papers we have counts for
7 papers
Where Words Begin and End
Husein Zolkepli
Forced aligners locate each word of a known transcript in the audio, and many speech pipelines depend on them. In our own data, a widely used CTC aligner inserts a pause at nearly…
How to Reduce Whisper Hallucination
Husein Zolkepli
Whisper is still what runs in production: one permissively licensed checkpoint, 99 languages, no per-language tuning. But it writes sentences nobody said. On 42 clips of pure room…
Improving X-Codec-2.0 for Multi-Lingual Speech: 25 Hz Latent Rate and 24 kHz Sampling
Husein Zolkepli
X-Codec-2.0 has shown strong performance in neural audio compression and multilingual speech modeling, operating at a 50 Hz latent rate and a 16 kHz sampling rate using frozen HuBE…
MMMModal -- Multi-Images Multi-Audio Multi-turn Multi-Modal
Husein Zolkepli, Aisyah Razak, Kamarul Adha +1
Our contribution introduces a groundbreaking multimodal large language model designed to comprehend multi-images, multi-audio, and multi-images-multi-audio within a single multitur…
Multi-Lingual Malaysian Embedding: Leveraging Large Language Models for Semantic Representations
Husein Zolkepli, Aisyah Razak, Kamarul Adha +1
In this work, we present a comprehensive exploration of finetuning Malaysian language models, specifically Llama2 and Mistral, on embedding tasks involving negative and positive pa…
Large Malaysian Language Model Based on Mistral for Enhanced Local Language Understanding
Husein Zolkepli, Aisyah Razak, Kamarul Adha +1
In this paper, we present significant advancements in the pretraining of Mistral 7B, a large-scale language model, using a dataset of 32.6 GB, equivalent to 1.1 billion tokens. We…
MaLLaM -- Malaysia Large Language Model
Husein Zolkepli, Aisyah Razak, Kamarul Adha +1
Addressing the gap in Large Language Model pretrained from scratch with Malaysian context, We trained models with 1.1 billion, 3 billion, and 5 billion parameters on a substantial…