Tootfinder

Zero-shot Context Biasing with Trie-based Decoding using Synthetic Multi-Pronunciation
Changsong Liu, Yizhou Peng, Eng Siong Chng
https://arxiv.org/abs/2508.17796 https://

Zero-shot Context Biasing with Trie-based Decoding using Synthetic Multi-Pronunciation
Contextual automatic speech recognition (ASR) systems allow for recognizing out-of-vocabulary (OOV) words, such as named entities or rare words. However, it remains challenging due to limited training data and ambiguous or inconsistent pronunciations. In this paper, we propose a synthesis-driven multi-pronunciation contextual biasing method that performs zero-shot contextual ASR on a pretrained Whisper model. Specifically, we leverage text-to-speech (TTS) systems to synthesize diverse speech sa…

The dude writing Colorado State's official press guide pronunciation table's got jokes. 😁

#coloradostate #humor #football

I apologize for bringing even an oblique mention of That Turd into your TL unnecessarily, but I just want to say that “HEPATITAS” (with the obligatory middle school humor pronunciation) is a magnificently embarrassing typo, and I hope people are cat-calling it at his memory long after he’s dead.

Good Morning #Canada
After recently moving to Belle Ewart (Bell You-wart), I find myself making strange pronunciations of my town to see people's reactions. Life is short... take pleasure where you can. But if you're a Canadian living in one of the towns mentioned in this CBC article, do you proudly tell people where you live, or do you mumble the answer and hope no one responds. I think you embrace the fun and show pride in your community by investing in t-shirts and hats emblazoned with your town name. But that's just me - some old retired guy with no reputation worth saving.
BTW, if you're looking for a conversation starter, here's a Geographic Fact - the distance between Dildo Newfoundland and Climax Saskatchewan is 5,551 km, about 2 days and 16 hours by car.
#CanadaIsAwesome #CivicPride
https://www.cbc.ca/television/stillstanding/the-most-hilarious-town-names-in-canada-1.5440187

Towards stable AI systems for Evaluating Arabic Pronunciations
Hadi Zaatiti, Hatem Hajri, Osama Abdullah, Nader Masmoudi
https://arxiv.org/abs/2508.19587 https://

Towards stable AI systems for Evaluating Arabic Pronunciations
Modern Arabic ASR systems such as wav2vec 2.0 excel at word- and sentence-level transcription, yet struggle to classify isolated letters. In this study, we show that this phoneme-level task, crucial for language learning, speech therapy, and phonetic research, is challenging because isolated letters lack co-articulatory cues, provide no lexical context, and last only a few hundred milliseconds. Recogniser systems must therefore rely solely on variable acoustic cues, a difficulty heightened by A…

Read to Hear: A Zero-Shot Pronunciation Assessment Using Textual Descriptions and LLMs
Yu-Wen Chen, Melody Ma, Julia Hirschberg
https://arxiv.org/abs/2509.14187 https://

Read to Hear: A Zero-Shot Pronunciation Assessment Using Textual Descriptions and LLMs
Automatic pronunciation assessment is typically performed by acoustic models trained on audio-score pairs. Although effective, these systems provide only numerical scores, without the information needed to help learners understand their errors. Meanwhile, large language models (LLMs) have proven effective in supporting language learning, but their potential for assessing pronunciation remains unexplored. In this work, we introduce TextPA, a zero-shot, Textual description-based Pronunciation Ass…

I hadn't had a drink in ages and don't particularly feel like changing this. But otherwise this could be an amazing self-research project for my Spanish learning!
«The 2025 #IgNobel Peace Prize went to researchers who proved a popular myth: a small dose of #alcohol can improve foreign language pronunciation!»
https://mstdn.science/@serhii/115241769055871087

PAC: Pronunciation-Aware Contextualized Large Language Model-based Automatic Speech Recognition
Li Fu, Yu Xin, Sunlu Zeng, Lu Fan, Youzheng Wu, Xiaodong He
https://arxiv.org/abs/2509.12647

PAC: Pronunciation-Aware Contextualized Large Language Model-based Automatic Speech Recognition
This paper presents a Pronunciation-Aware Contextualized (PAC) framework to address two key challenges in Large Language Model (LLM)-based Automatic Speech Recognition (ASR) systems: effective pronunciation modeling and robust homophone discrimination. Both are essential for raw or long-tail word recognition. The proposed approach adopts a two-stage learning paradigm. First, we introduce a pronunciation-guided context learning method. It employs an interleaved grapheme-phoneme context modeling …

Fine-Tuning Large Multimodal Models for Automatic Pronunciation Assessment
Ke Wang, Wenning Wei, Yan Deng, Lei He, Sheng Zhao
https://arxiv.org/abs/2509.15701 https://

Fine-Tuning Large Multimodal Models for Automatic Pronunciation Assessment
Automatic Pronunciation Assessment (APA) is critical for Computer-Assisted Language Learning (CALL), requiring evaluation across multiple granularities and aspects. Large Multimodal Models (LMMs) present new opportunities for APA, but their effectiveness in fine-grained assessment remains uncertain. This work investigates fine-tuning LMMs for APA using the Speechocean762 dataset and a private corpus. Fine-tuning significantly outperforms zero-shot settings and achieves competitive results on si…

UtterTune: LoRA-Based Target-Language Pronunciation Edit and Control in Multilingual Text-to-Speech
Shuhei Kato
https://arxiv.org/abs/2508.09767 https://ar…

UtterTune: LoRA-Based Target-Language Pronunciation Edit and Control in Multilingual Text-to-Speech
We propose UtterTune, a lightweight adaptation method that fine-tunes a multilingual text-to-speech (TTS) system based on a large language model (LLM) architecture, designed to enhance the controllability of pronunciation in a target language while preserving performance in others. While LLM architectures have enabled TTS models to achieve remarkable naturalness, accurately modeling grapheme-to-phoneme (G2P) mapping and prosody remains challenging, especially when the model omits an explicit G2…