Support Linguist – Text to Speech (TTS)

Covalen

Sandyford

Hybrid

EUR 42,000 - 56,000

Full time

7 days ago

Be an early applicant

Get a reply from this employer

Your resume and cover letter, tailored to exactly what they’re hiring for.

Get past ATS filters

Benefits offered by this job

Fully subsidised healthcare

25 days annual leave

Employee assistance programme (EAP)

PRSA Scheme

Extensive Training & Development

Annual company events

Bike to work scheme

Free parking on-site

Job summary

Covalen is seeking a detail-oriented Support Linguist for Text-to-Speech to enhance speech technologies in the candidate language. You will evaluate linguistic and audio outputs, contribute to text normalization, and identify language-specific challenges with data-driven approaches.

The role involves responsibilities before and after launch, including regression testing, dataset curation, and G2P evaluation. Collaboration and independent work are expected.

Qualifications

Degree in Linguistics or demonstrated IPA knowledge.
Working knowledge of Pandas for data review and manipulation.
Regex for text processing and validation.
Native or near-native speaker - Italian, Spanish, or Portuguese-Brazilian with fluent English (C1).
Experience in speech/language technologies is advantageous.

Responsibilities

Evaluate linguistic and audio quality of Text-to-Speech outputs.
Support improvements in the TTS pipeline including text normalization and pronunciation prediction.
Identify language-specific challenges and propose test scenarios.
Perform data quality checks and improve linguistic datasets.
Analyse text and linguistic datasets to identify patterns and errors.
Use Pandas and Regex for linguistic data analysis and validation.

Skills

Linguistic knowledge

Attention to detail

Analytical thinking

Data interpretation

Communication skills

Team collaboration

Python basics

Education

Linguistics degree or IPA knowledge

Fluent English (C1)

Tools

Pandas

Regex

Python

G2P datasets

Job description

We are looking for detail orientated candidates for the role of Support Linguist – Text to Speech (TTS) to support the improvement of speech technologies by evaluating linguistic and audio quality, contributing to text normalization processes, and identifying language-specific challenges.

The candidate must possess strong linguistic knowledge, excellent attention to detail, and strong analytical and communication skills. They will be flexible and analytical with the ability to understand language structures, interpret datasets, and apply predefined guidelines to ensure accuracy and quality of output. Experience working with Pandas, regular expressions (Regex), Python, or similar data and text-processing tools is particularly valuable, as these are relevant to the day-to-day activities of the role.

Duties and Responsibilities

As a Support Linguist for Text-to-Speech, you will contribute to improving the quality and performance of speech technologies in your language of expertise.

Responsibilities include, but are not limited to:

Evaluate linguistic and audio quality of Text-to-Speech outputs.
Support improvements in the TTS pipeline, including text normalization and pronunciation prediction.
Identify language-specific challenges and propose test scenarios to evaluate them.
Perform data quality checks and contribute to improving linguistic datasets.
Analyse text and linguistic datasets to identify patterns, errors, and opportunities for improvement.
Use text-processing and data-analysis techniques to support linguistic investigations and quality checks.

Prior to Launch

Create and maintain regression tests for each locale within the language.
Develop and maintain text normalization test sets.
Source and review datasets used to train text normalization systems.
Develop text normalization rules to ensure accuracy against defined test sets.
Create and maintain pronunciation datasets used for grapheme-to-phoneme (G2P) evaluation.
Identify and resolve language-specific challenges such as grammatical gender, word stress, segmentation, tone prediction, and inflection.
Perform targeted data quality checks and audio evaluations.
Use structured data and text-processing techniques, including Pandas and regular expressions (Regex) where applicable, to analyse and validate linguistic data.

After Launch

Investigate and resolve language-related issues reported through the Linguist Runbook.
Continue improving text normalization and linguistic rules.
Ensure voice deployments pass capability testing as part of the launch review process.
Perform ongoing audio evaluations to ensure consistent quality.

We're seeking individuals from diverse backgrounds with a passion for language and technology.

Essential competencies needed for this role are:

High attention to detail and strong analytical thinking.
Ability to evaluate linguistic and speech outputs objectively.
Strong spoken comprehension and ability to assess audio quality.
Ability to work with structured datasets and follow testing guidelines.
Strong collaboration and communication skills.
Comfortable working with digital tools and data workflows.
Flexible and adaptable in a fast-paced environment.
Ability to work independently and collaboratively while meeting project deadlines.

Requirements:

Degree in Linguistics or a related field or demonstrated knowledge of the International Phonetic Alphabet (IPA).
Working knowledge of Pandas for reviewing, filtering, manipulating, and analysing structured datasets.
Working knowledge of regular expressions (Regex) for text processing, pattern matching, and validation.
Native or near-native speaker - Italian or Spanish or Portuguese-Brazilian with fluent English (C1)
Computational experience or strong interest in speech and language technologies, including working with structured text datasets and text-processing tools.
Experience with linguistic analysis, data annotation, or speech technologies is advantageous.

Additional skills that are advantageous: