6 citations · 19 across the 12 of their papers we have counts for
15 papers
Speculative Speech Recognition by Audio-Prefixed Low-Rank Adaptation of Language Models
Bolaji Yusuf, Murali Karthick Baskar, Andrew Rosenberg +1
This paper explores speculative speech recognition (SSR), where we empower conventional automatic speech recognition (ASR) with speculation capabilities, allowing the recognizer to…
Speech Prefix-Tuning with RNNT Loss for Improving LLM Predictions
Murali Karthick Baskar, Andrew Rosenberg, Bhuvana Ramabhadran +2
In this paper, we focus on addressing the constraints faced when applying LLMs to ASR. Recent works utilize prefixLM-type models, which directly apply speech as a prefix to LLMs fo…
O-1: Self-training with Oracle and 1-best Hypothesis
Murali Karthick Baskar, Andrew Rosenberg, Bhuvana Ramabhadran +1
We introduce O-1, a new self-training objective to reduce training bias and unify training and evaluation metrics for speech recognition. O-1 is a faster variant of Expected Minimu…
Robust Knowledge Distillation from RNN-T Models With Noisy Training Labels Using Full-Sum Loss
Mohammad Zeineldeen, Kartik Audhkhasi, Murali Karthick Baskar +1
This work studies knowledge distillation (KD) and addresses its constraints for recurrent neural network transducer (RNN-T) models. In hard distillation, a teacher model transcribe…
Speaker adaptation for Wav2vec2 based dysarthric ASR
Murali Karthick Baskar, Tim Herzig, Diana Nguyen +4
Dysarthric speech recognition has posed major challenges due to lack of training data and heavy mismatch in speaker characteristics. Recent ASR systems have benefited from readily…
Ask2Mask: Guided Data Selection for Masked Speech Modeling
Murali Karthick Baskar, Andrew Rosenberg, Bhuvana Ramabhadran +2
Masked speech modeling (MSM) methods such as wav2vec2 or w2v-BERT learn representations over speech frames which are randomly masked within an utterance. While these methods improv…
EAT: Enhanced ASR-TTS for Self-supervised Speech Recognition
Murali Karthick Baskar, Lukáš Burget, Shinji Watanabe +2
Self-supervised ASR-TTS models suffer in out-of-domain data conditions. Here we propose an enhanced ASR-TTS (EAT) model that incorporates two main features: 1) The ASR…
BUT Opensat 2019 Speech Recognition System
Martin Karafiát, Murali Karthick Baskar, Igor Szöke +3
The paper describes the BUT Automatic Speech Recognition (ASR) systems submitted for OpenSAT evaluations under two domain categories such as low resourced languages and public safe…
Semi-supervised Sequence-to-sequence ASR using Unpaired Speech and Text
Murali Karthick Baskar, Shinji Watanabe, Ramon Astudillo +3
Sequence-to-sequence automatic speech recognition (ASR) models require large quantities of data to attain high performance. For this reason, there has been a recent surge in intere…
Analysis of Multilingual Sequence-to-Sequence speech recognition systems
Martin Karafiát, Murali Karthick Baskar, Shinji Watanabe +3
This paper investigates the applications of various multilingual approaches developed in conventional hidden Markov model (HMM) systems to sequence-to-sequence (seq2seq) automatic…
Promising Accurate Prefix Boosting for sequence-to-sequence ASR
Murali Karthick Baskar, Lukáš Burget, Shinji Watanabe +3
In this paper, we present promising accurate prefix boosting (PAPB), a discriminative training technique for attention based sequence-to-sequence (seq2seq) ASR. PAPB is devised to…
Language model integration based on memory control for sequence to sequence speech recognition
Jaejin Cho, Shinji Watanabe, Takaaki Hori +4
In this paper, we explore several new schemes to train a seq2seq model to integrate a pre-trained LM. Our proposed fusion methods focus on the memory cell state and the hidden stat…
Transfer learning of language-independent end-to-end ASR with language model fusion
Hirofumi Inaguma, Jaejin Cho, Murali Karthick Baskar +2
This work explores better adaptation methods to low-resource languages using an external language model (LM) under the framework of transfer learning. We first build a language-ind…
Multilingual sequence-to-sequence speech recognition: architecture, transfer learning, and language modeling
Jaejin Cho, Murali Karthick Baskar, Ruizhi Li +6
Sequence-to-sequence (seq2seq) approach for low-resource ASR is a relatively new direction in speech research. The approach benefits by performing model training without using lexi…
Residual Memory Networks: Feed-forward approach to learn long temporal dependencies
Murali Karthick Baskar, Martin Karafiat, Lukas Burget +3
Training deep recurrent neural network (RNN) architectures is complicated due to the increased network complexity. This disrupts the learning of higher order abstracts using deep R…