User-friendly Automatic Transcription of Low-resource Languages: Plugging ESPnet into Elpis

This paper reports on progress integrating the speech recognition toolkit ESPnet into Elpis, a web front-end originally designed to provide access to the Kaldi automatic speech recognition toolkit. The goal of this work is to make end-to-end speech recognition models available to language workers via a user-friendly graphical interface. Encouraging results are reported on (i) development of an ESPnet recipe for use in Elpis, with preliminary results on data sets previously used for training acoustic models with the Persephone toolkit along with a new data set that had not previously been used in speech recognition, and (ii) incorporating ESPnet into Elpis along with UI enhancements and a CUDA-supported Dockerfile.

• Publication year

  2020

• Venue

  COMPUTEL

• Publication date

  2020-12-14

• Fields of study

  Linguistics, Computer Science, Engineering

• Identifiers
  DOI 10.33011/COMPUTEL.V1I.969arXiv 2101.03027
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations

  2020cited by this paper
• Don’t Stop Pretraining: Adapt Language Models to Domains and Tasks

  2020cited by this paper
• A Survey on Contextual Embeddings

  2020cited by this paper
• Universal Phone Recognition with a Multilingual Allophone System

  2020influential reference
• Unsupervised Pretraining Transfers Well Across Languages

  2020cited by this paper
• The Rwth Asr System for Ted-Lium Release 2: Improving Hybrid Hmm With Specaugment

  2020cited by this paper
• Towards a Speech Recognizer for Komi, an Endangered and Low-Resource Uralic Language

  2020cited by this paper
• Response from the editor to "toward open data policies in phonetics: what we can gain and how we can avoid pitfalls"

  2020cited by this paper
• The Blind Spots of Digital Innovation Fetishism

  2020cited by this paper
• Kaldi-Web: An Installation-Free, On-Device Speech Recognition System

  2020cited by this paper
• Phonemic Transcription of Low-Resource Languages: To What Extent can Preprocessing be Automated?

  2020cited by this paper
• Elpis, an Accessible Speech-to-Text Tool

  2019cited by this paper
• Future Directions in Technological Support for Language Documentation

  2019cited by this paper
• Massively Multilingual Adversarial Speech Recognition

  2019cited by this paper
• BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

  2019cited by this paper
• PyTorch: An Imperative Style, High-Performance Deep Learning Library

  2019cited by this paper
• Multilingual Dependency Parsing for Low-Resource Languages: Case Studies on North Saami and Komi-Zyrian

  2018cited by this paper
• Deep Contextualized Word Representations

  2018cited by this paper
• Sequence-Based Multi-Lingual Low Resource Speech Recognition

  2018cited by this paper
• ESPnet: End-to-End Speech Processing Toolkit

  2018cited by this paper
• Improved training of end-to-end attention models for speech recognition

  2018cited by this paper
• End-to-end Speech Recognition Using Lattice-free MMI

  2018cited by this paper
• Adversarial Multilingual Training for Low-Resource Speech Recognition

  2018cited by this paper
• Integrating automatic transcription into the language documentation workflow: Experiments with Na data and the Persephone toolkit

  2018cited by this paper
• Transfer Learning of Language-independent End-to-end ASR with Language Model Fusion

  2018cited by this paper
• Building Speech Recognition Systems for Language Documentation: The CoEDL Endangered Language Pipeline and Inference System (Elpis)

  2018cited by this paper
• The Pytorch-kaldi Speech Recognition Toolkit

  2018cited by this paper
• Language Science Press business model: Evaluated version of the 2015 model

  2018cited by this paper
• Evaluation Phonemic Transcription of Low-Resource Tonal Languages for Language Documentation

  2018cited by this paper
• Language independent end-to-end architecture for joint language identification and speech recognition

  2017cited by this paper
• Multilingual processing of speech via web services

  2017cited by this paper
• Japhug

  2017cited by this paper
• State-of-the-Art Speech Recognition with Sequence-to-Sequence Models

  2017cited by this paper
• Attention is All you Need

  2017cited by this paper
• Multilingual Speech Recognition with a Single End-to-End Model

  2017cited by this paper
• Building an ASR System for a Low-resource Language Through the Adaptation of a High-resource Language ASR System: Preliminary Results

  2017cited by this paper
• Hybrid CTC/Attention Architecture for End-to-End Speech Recognition

  2017cited by this paper
• Phonemic and Graphemic Multilingual CTC Based Speech Recognition

  2017cited by this paper
• Endangered Language Documentation: Bootstrapping a Chatino Speech Corpus, Forced Aligner, ASR

  2016cited by this paper
• Language Documentation meets Language Technology

  2015cited by this paper
• Documenting and Researching Endangered Lan- guages: The Pangloss Collection

  2014cited by this paper
• Using out-of-language data to improve an under-resourced speech recognizer

  2014cited by this paper
• Cross-Lingual Phone Mapping for Large Vocabulary Speech Recognition of Under-Resourced Languages

  2014cited by this paper
• GloVe: Global Vectors for Word Representation

  2014cited by this paper
• Deep Speech: Scaling up end-to-end speech recognition

  2014cited by this paper
• Automatic speech recognition for under-resourced languages: A survey

  2014cited by this paper
• Finding a way into a family of tone languages: The story and methods of the Chatino Language Documentation Project

  2014cited by this paper
• Distributed Representations of Words and Phrases and their Compositionality

  2013cited by this paper
• Multilingual acoustic models using distributed deep neural networks

  2013cited by this paper
• Multilingual MLP features for low-resource LVCSR systems

  2012cited by this paper
• Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups

  2012cited by this paper
• ADADELTA: An Adaptive Learning Rate Method

  2012cited by this paper
• Phonology, tone and the functions of tone in San Juan Quiahije Chatino

  2011cited by this paper
• Cross-lingual portability of Chinese and english neural network features for French and German LVCSR

  2011cited by this paper
• The Kaldi Speech Recognition Toolkit

  2011cited by this paper
• Cross-lingual portability of MLP-based tandem features - a case study for English and Hungarian

  2008cited by this paper
• Cross-Domain and Cross-Language Portability of Acoustic Features Estimated by Multilayer Perceptrons

  2006cited by this paper
• Chapter 1 Language documentation: What is it and what is it good for?

  2006cited by this paper
• First steps in fast acoustic modeling for a new target language: application to Vietnamese

  2005cited by this paper
• Experiments on cross-language acoustic modeling

  2001cited by this paper
• State-of-the-art speech recognition u.s. research and business update

  1987cited by this paper

• Evaluating Speech Foundation Models for Automatic Speech Recognition in the Low-Resource Kanyen'kéha Language

  2025cites this paper
• Supporting SENCOTEN Language Documentation Efforts with Automatic Speech Recognition

  2025cites this paper
• Breaking the Transcription Bottleneck: Fine-tuning ASR Models for Extremely Low-Resource Fieldwork Languages

  2025cites this paper
• Access Control Framework for Language Collections

  2024cites this paper
• Now You See Me, Now You Don’t: ‘Poverty of the Stimulus’ Problems and Arbitrary Correspondences in End-to-End Speech Models

  2024cites this paper
• Speech-to-text recognition for multilingual spoken data in language documentation

  2023cites this paper
• Application of Speech Processes for the Documentation of Kréyòl Gwadloupéyen

  2023cites this paper
• Learnings from Technological Interventions in a Low Resource Language: Enhancing Information Access in Gondi

  2022cites this paper
• Phoneme transcription of endangered languages: an evaluation of recent ASR architectures in the single speaker scenario

  2022cites this paper
• End-to-End Automatic Speech Recognition: Its Impact on the Workflowin Documenting Yoloxóchitl Mixtec

  2021cites this paper
• Keyword spotting for audiovisual archival search in Uralic languages

  2021cites this paper
• Spoken Term Detection Methods for Sparse Transcription in Very Low-resource Settings

  2021cites this paper
• The Relevance of the Source Language in Transfer Learning for ASR

  2021cites this paper