Blog

Explore the blog

Browse practical insights on transcription, captioning, and speech data — with the newest posts first.

Showing 12 posts on this page (527 total)

Crowdsourced Speech Data: A Cornerstone of Dataset Acquisition featured image

Crowdsourced Speech Data: A Cornerstone of Dataset Acquisition

By Way With Words Team

This article explores the benefits of crowdsourced speech data collection, the platforms that enable it, dataset quality, and the ethical considerations involved.

Read article
Scripted vs Unscripted Speech: Analysing Benefits and Challenges featured image

Scripted vs Unscripted Speech: Analysing Benefits and Challenges

By Way With Words Team

This article explores the differences between scripted vs unscripted speech, their advantages, limitations, and practical applications in today’s AI landscape.

Read article
How Do You Test for Fairness in Speech AI? featured image

How Do You Test for Fairness in Speech AI?

By Way With Words Team

How Do You Test for Fairness in Multilingual Speech Data? Building Equitable Systems that Function for Everyone Artificial intelligence has rapidly become...

Read article
What Multilingual Open Speech Corpora Exist for Research? featured image

What Multilingual Open Speech Corpora Exist for Research?

By Way With Words Team

What Multilingual Speech Corpora Exist for Open Research? Why is Access to Open Multilingual Speech Datasets Important? Artificial intelligence systems tha...

Read article
Synthetic Dialect Generation: Training Machine Learning Models featured image

Synthetic Dialect Generation: Training Machine Learning Models

By Way With Words Team

Machine learning has made remarkable progress in synthetic dialect generation, opening new opportunities for speech synthesis, data inclusivity, and language preservation.

Read article
Best Practices for Tagging Multilingual Code-mixing in Audio featured image

Best Practices for Tagging Multilingual Code-mixing in Audio

By Way With Words Team

Tagging multilingual code-mixing in audio files is one of the most complex but rewarding tasks in speech annotation.

Read article
How Loanwords Impact Automatic Speech Recognition Models featured image

How Loanwords Impact Automatic Speech Recognition Models

By Way With Words Team

What is the Impact of Loanwords on ASR Models? Understanding How Loanwords Affect Automatic Speech Recognition Automatic Speech Recognition (ASR) has becom...

Read article
Why Is Phonetic Transcription Useful in Multilingual Datasets? featured image

Why Is Phonetic Transcription Useful in Multilingual Datasets?

By Way With Words Team

This article explores the foundations of phonetic transcription, its benefits in multilingual ASR and TTS systems.

Read article
Effectively Manage Multilingual ASR Training Pipelines at Scale featured image

Effectively Manage Multilingual ASR Training Pipelines at Scale

By Way With Words Team

This article explores the key components of multilingual ASR training pipelines, and outlines the strategies and tools that engineers and developers use to manage complex datasets at scale.

Read article
Avoid Common Speech Data Errors in Cross-Linguistic Corpora featured image

Avoid Common Speech Data Errors in Cross-Linguistic Corpora

By Way With Words Team

Understanding the most common types of speech data errors, their causes, and how they can be mitigated is essential for any team working on multilingual or cross-linguistic voice datasets.

Read article
Practical Steps to Preserve Cultural Context in Speech Data featured image

Practical Steps to Preserve Cultural Context in Speech Data

By Way With Words Team

Preserving cultural context in speech data has far-reaching effects, particularly in localisation and conversational AI.

Read article
Why is Tonal Language Essential in African and Asian Speech Data? featured image

Why is Tonal Language Essential in African and Asian Speech Data?

By Way With Words Team

This article explores the essence of tonal languages, the challenges of recording them accurately and the wide-ranging applications of tone-aware ASR models.

Read article