Publications (14)
OWSM v3.1: Better and Faster Open Whisper-Style Speech Models based on E-Branchformer
Yifan Peng, Jinchuan Tian, William Chen +9
Recent studies have highlighted the importance of fully open foundation models. The Open Whisper-style Speech Model (OWSM) is an initial step towards reproducing OpenAI Whisper usi…
Joint Optimization of Streaming and Non-Streaming Automatic Speech Recognition with Multi-Decoder and Knowledge Distillation
Muhammad Shakeel, Yui Sudo, Yifan Peng +1
End-to-end (E2E) automatic speech recognition (ASR) can operate in two modes: streaming and non-streaming, each with its pros and cons. Streaming ASR processes the speech frames in…
Joint Beam Search Integrating CTC, Attention, and Transducer Decoders
Yui Sudo, Muhammad Shakeel, Yosuke Fukumoto +4
End-to-end automatic speech recognition (E2E-ASR) can be classified by its decoder architectures, such as connectionist temporal classification (CTC), recurrent neural network tran…
Benford's laws tests on S&P500 daily closing values and the corresponding daily log-returns both point to huge non-conformity
Marcel Ausloos, Valerio Ficcadenti, Gurjeet Dhesi +1
The so-called Benford's laws are of frequent use in order to observe anomalies and regularities in data sets, in particular, in election results and financial statements. Yet, basi…
Contextualized Automatic Speech Recognition with Attention-Based Bias Phrase Boosted Beam Search
Yui Sudo, Muhammad Shakeel, Yosuke Fukumoto +2
End-to-end (E2E) automatic speech recognition (ASR) methods exhibit remarkable performance. However, since the performance of such methods is intrinsically linked to the context pr…
4D ASR: Joint modeling of CTC, Attention, Transducer, and Mask-Predict decoders
Yui Sudo, Muhammad Shakeel, Brian Yan +2
The network architecture of end-to-end (E2E) automatic speech recognition (ASR) can be classified into several models, including connectionist temporal classification (CTC), recurr…
Contextualized End-to-end Automatic Speech Recognition with Intermediate Biasing Loss
Muhammad Shakeel, Yui Sudo, Yifan Peng +1
Contextualized end-to-end automatic speech recognition has been an active research area, with recent efforts focusing on the implicit learning of contextual phrases based on the fi…
Metric-based multimodal meta-learning for human movement identification via footstep recognition
Muhammad Shakeel, Katsutoshi Itoyama, Kenji Nishida +1
We describe a novel metric-based learning approach that introduces a multimodal framework and uses deep audio and geophone encoders in siamese configuration to design an adaptable…
CALM: Joint Contextual Acoustic-Linguistic Modeling for Personalization of Multi-Speaker ASR
Muhammad Shakeel, Yosuke Fukumoto, Chikara Maeda +2
We present CALM, a joint Contextual Acoustic-Linguistic Modeling framework for multi-speaker automatic speech recognition (ASR). In personalized AI scenarios, the joint availabilit…
Unifying Diarization, Separation, and ASR with Multi-Speaker Encoder
Muhammad Shakeel, Yui Sudo, Yifan Peng +2
This paper presents a unified multi-speaker encoder (UME), a novel architecture that jointly learns representations for speaker diarization (SD), speech separation (SS), and multi-…
Contextualized Automatic Speech Recognition with Dynamic Vocabulary
Yui Sudo, Yosuke Fukumoto, Muhammad Shakeel +2
Deep biasing (DB) enhances the performance of end-to-end automatic speech recognition (E2E-ASR) models for rare words or contextual phrases using a bias list. However, most existin…
Reproducing Whisper-Style Training Using an Open-Source Toolkit and Publicly Available Data
Yifan Peng, Jinchuan Tian, Brian Yan +13
Pre-training speech models on large volumes of data has achieved remarkable success. OpenAI Whisper is a multilingual multitask model trained on 680k hours of supervised speech dat…
DYNAC: Dynamic Vocabulary based Non-Autoregressive Contextualization for Speech Recognition
Yui Sudo, Yosuke Fukumoto, Muhammad Shakeel +3
Contextual biasing (CB) improves automatic speech recognition for rare and unseen phrases. Recent studies have introduced dynamic vocabulary, which represents context phrases as ex…
OWSM-CTC: An Open Encoder-Only Speech Foundation Model for Speech Recognition, Translation, and Language Identification
Yifan Peng, Yui Sudo, Muhammad Shakeel +1
There has been an increasing interest in large speech models that can perform multiple tasks in a single model. Such models usually adopt an encoder-decoder or decoder-only archite…