Lithuanian Broadcast Speech Transcription Using Semi-supervised Acoustic Model Training

Lileikytė, Rasa; Gorin, Arseniy; Lamel, Lori; Gauvain, Jean‐Luc; Fraga-Silva, Thiago

doi:10.1016/j.procs.2016.04.037

Cited by 12 publications

(7 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They apply semi-supervised learning for lexicon learning as well as acoustic modeling. Similarly, Veselỳ et al (2018) & Lileikytė et al (2016 (Karita et al, 2019). In both cases, they assume that the encoded speech features in the current minibatch are sampled from one distribution and encoded text features in the current minibatch are sampled from a second distribution.…”

Section: Related Workmentioning

confidence: 99%

Improving End-to-End Bangla Speech Recognition with Semi-supervised Training

Sadeq

Chowdhury²,

Utshaw³

et al. 2020

Findings of the Association for Computational Linguistics: EMNLP 2020

View full text Add to dashboard Cite

Automatic speech recognition systems usually require large annotated speech corpus for training. The manual annotation of a large corpus is very difficult. It can be very helpful to use unsupervised and semi-supervised learning methods in addition to supervised learning. In this work, we focus on using a semi-supervised training approach for Bangla Speech Recognition that can exploit large unpaired audio and text data. We encode speech and text data in an intermediate domain and propose a novel loss function based on the global encoding distance between encoded data to guide the semisupervised training. Our proposed method reduces the Word Error Rate (WER) of the system from 37% to 31.9%.

show abstract

Section: Related Workmentioning

confidence: 99%

Improving End-to-End Bangla Speech Recognition with Semi-supervised Training

Sadeq

Chowdhury²,

Utshaw³

et al. 2020

Findings of the Association for Computational Linguistics: EMNLP 2020

View full text Add to dashboard Cite

show abstract

“…Nonetheless, important researches are carried out in the field of annotation of speech, which is the most labour and time consuming activity. The majority of investigators demonstrate abilities in transcription of broadcast speech (Esteve et al, 2010;Lileikytė et al, 2016;Mansikkaniemi et al, 2017), extent of annotation and linkage (Johannessen et al, 2007).…”

Section: Related Workmentioning

confidence: 99%

Lithuanian Speech Corpus Liepa for Development of Human-Computer Interfaces Working in Voice Recognition and Synthesis Mode

Laurinčiukaitė¹,

Telksnys²,

Kasparaitis³

et al. 2018

Informatica

View full text Add to dashboard Cite

The problem of speech corpus for design of human-computer interfaces working in voice recognition and synthesis mode is investigated. Specific requirements of speech corpus for speech recognizers and synthesizers were accented. It has been discussed that in order to develop above mentioned speech corpus, it has to consist of two parts. One part of speech corpus should be presented for the needs of Lithuanian text-to-speech synthesizers, another part of speech corpus-for the needs of Lithuanian speech recognition engines. It has been determined that the part of speech corpus designed for speech recognition engines has to ensure the availability to present language specificity by the use of different sets of phonemes. According to the research results, the speech corpus Liepa, which consists of two parts, was developed. This speech corpus opens possibilities for cost-effective and flexible development of human-computer interfaces working in voice recognition and synthesis mode.

show abstract

“…Semi-supervised acoustic model training techniques, in which a small amount of reference data with annotations is used to bootstrap an ASR system that provides training labels for untranscribed speech data, have been researched intensively and various training strategies and data selection criteria have been proposed [26,27,28,29,30,31,32,33,34]. These techniques are especially useful for increasing the available training data while building ASR systems for under-resourced languages with minimal amount of manually annotated data.…”

Section: Introductionmentioning

confidence: 99%

Semi-supervised acoustic model training for speech with code-switching

Yılmaz

McLaren

Heuvel

et al. 2018

Speech Communication

View full text Add to dashboard Cite

In the FAME! project, we aim to develop an automatic speech recognition (ASR) system for Frisian-Dutch code-switching (CS) speech extracted from the archives of a local broadcaster with the ultimate goal of building a spoken document retrieval system. Unlike Dutch, Frisian is a low-resourced language with a very limited amount of manually annotated speech data. In this paper, we describe several automatic annotation approaches to enable using of a large amount of raw bilingual broadcast data for acoustic model training in a semi-supervised setting. Previously, it has been shown that the best-performing ASR system is obtained by two-stage multilingual deep neural network (DNN) training using 11 hours of manually annotated CS speech (reference) data together with speech data from other high-resourced languages. We compare the quality of transcriptions provided by this bilingual ASR system with several other approaches that use a language recognition system for assigning language labels to raw speech segments at the front-end and using monolingual ASR resources for transcription. We further investigate automatic annotation of the speakers appearing in the raw broadcast data by first labeling with (pseudo) speaker tags using a speaker diarization system and then linking to the known speakers appearing in the reference data using a speaker recognition system. These speaker labels are essential for speaker-adaptive training in the proposed setting. We train acoustic models using the manually and automatically annotated data and run recognition experiments on the development and test data of the FAME! speech corpus to quantify the quality of the automatic annotations. The ASR and CS detection results demonstrate the potential of using automatic language and speaker tagging in semi-supervised bilingual acoustic model training.

show abstract

Lithuanian Broadcast Speech Transcription Using Semi-supervised Acoustic Model Training

Cited by 12 publications

References 21 publications

Improving End-to-End Bangla Speech Recognition with Semi-supervised Training

Improving End-to-End Bangla Speech Recognition with Semi-supervised Training

Lithuanian Speech Corpus Liepa for Development of Human-Computer Interfaces Working in Voice Recognition and Synthesis Mode

Semi-supervised acoustic model training for speech with code-switching

Contact Info

Product

Resources

About