Fine-tuning the non-blank predictor of a factorized transducer on target-domain text improves recognition on that domain but degrades performance on the source domain. A single interpolation weight be...
arXiv:2604.06487v2 Announce Type: replace Abstract: Conventional end-to-end automatic speech recognition (ASR) systems rely on paired speech-text data for domain adaptation. Recent LLM-based ASR arch...
arXiv:2609.14991v1 Announce Type: new Abstract: Open Thai automatic speech recognition (ASR) is dominated by offline, Whisper-based models that read the whole utterance before transcribing, ruling ou...
SPAR-K is a scheduled periodic alternating early‑exit framework for interleaved spoken language models that reduces decoding depth for speech tokens while maintaining quality. It lets most speech positions exit at a fixed intermediate layer and inserts periodic full‑depth refresh steps to counter distribution shift. Experiments on Step‑Audio‑2‑mini and GLM‑4‑Voice show up to 11 % depth reduction with less than 0.82 % drop in question‑answering accuracy and negligible impact on MOS and WER.
The paper introduces asymmetric classifier‑free guidance (CFG) for target‑speaker ASR using Whisper, where a speaker‑conditioned branch predicts the target transcript and a speaker‑unconditioned branch predicts serialized multi‑speaker transcripts. CFG modulates the influence of speaker conditioning during decoding via a single guidance scale, which is first set globally on development data and then refined per utterance by a lightweight encoder‑based predictor while keeping the recognition model fixed. The resulting system yields up to 21.8% relative WER reduction over a condition‑only baseline and 5.6% over standard conditional decoding under domain shifts.
arXiv:2609.22214v1 Announce Type: new Abstract: Long multilingual conversational spoken question answering requires systems to balance long-range transcript semantics with sparse acoustic and speaker...
The paper introduces a new benchmark of 13,044 spontaneous Nepali‑English code‑switched podcast utterances, totaling 15.45 h, and evaluates a two‑stage synthetic‑to‑real adaptation framework for ASR. Stage 1 uses supervised fine‑tuning on 0.95 h of synthetic speech, reducing Mixed Error Rate from 23.40 % to 21.52 % and switch‑point error rate from 35.89 % to 28.77 %. Stage 2 applies Group Relative Policy Optimization with script and transcription rewards, yielding modest but statistically significant further improvements, particularly in English script confusion and switch‑point accuracy. The study also shows that an auxiliary script reward term offers no benefit beyond MER‑only optimization in this setting, and that the synthetic SFT provides the main cross‑domain transfer to spontaneous human speech.
Open Thai automatic speech recognition (ASR) is dominated by offline, Whisper-based models that read the whole utterance before transcribing, ruling out low-latency uses such as live captioning and vo...
Confucius4‑R2T2, a stable‑prefix streaming ASR model, is introduced to ensure that once a token is emitted it is never revised, thereby reducing wasted work and incorrect actions in real‑time transcription. The model is derived by post‑training Qwen3‑ASR‑1.7B on 1,500 hours of Chinese and English audio while keeping its audio encoder frozen, and it leverages a new concept called Longest Stable Prefix (LSP) to convert a full‑context model into a stable‑prefix streaming one without architectural changes. Evaluation shows that R2T2 achieves near full‑context quality with low latency and stable output, and its benefits generalize beyond the languages used in training.
arXiv:2509.10452v3 Announce Type: replace-cross Abstract: Pretrained automatic speech recognition (ASR) models such as Whisper perform well but still need domain adaptation to handle unseen parlance....
arXiv:2606. 16456v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) models enable efficient scaling, but training them from scratch remains prohibitively expensive.
arXiv:2609.17981v1 Announce Type: cross Abstract: Speech Large Language Models (Speech-LLMs), typically built from a pre-trained speech encoder, a modality projector, and an LLM fine-tuned with Low-R...