Jobs / Multimodal

Kinyarwanda Speakers: Audio Transcription for Speech Recognition

We are running an ongoing project to improve speech recognition models through accurate audio transcription… Pay as listed: $4/hr. Remote (United States).

Pay (as listed on Terac)
$4/hr · Hourly
Location
Remote · United States
Languages
English
Category
Multimodal
Posted
Nov 13, 2026

Skills

  • Kinyarwanda
  • Audio Transcription
  • Data Annotation
  • Speech Recognition

Job description

What We're Researching

We're running a paid project focused on training advanced speech recognition models for Kinyarwanda. High-quality transcription data is critical for building reliable voice technologies that understand diverse languages. Your work will directly improve how these systems process and recognize spoken Kinyarwanda in real-world applications.

How It Works

You will listen to short audio recordings and type out the spoken words with high accuracy. The work also involves adjusting time segmentations to ensure the text aligns perfectly with the audio track. This is an ongoing, remote opportunity requiring a commitment of at least 10 hours per week. All tasks are completed asynchronously through our specialized annotation platform.

Who This Is For

We are hiring fluent Kinyarwanda speakers who have strong listening and typing skills. Previous experience in transcription, translation, or data annotation is highly valued but not strictly required. You must be comfortable working independently and have a reliable internet connection to handle audio files.

Responsibilities

  • Listen to Kinyarwanda audio clips and transcribe the spoken content accurately
  • Adjust audio segmentations to align text boundaries with the spoken words
  • Commit at least 10 hours per week to complete ongoing tasks
  • Review transcripts to ensure high quality and proper spelling

Requirements

  • Fluent or native proficiency in Kinyarwanda
  • Strong written communication and typing skills
  • Access to a reliable computer and internet connection
  • Ability to dedicate 10 or more hours per week to the project

Who can apply

  • Located in Rwanda, Uganda, Democratic Republic of the Congo, or the United States
  • Must be 18 years or older
  • Native or mother-tongue proficiency in Kinyarwanda
  • Native, bilingual, or full professional working proficiency in English
  • Regular or occasional hands-on experience transcribing and editing audio files
  • Must have a dedicated quiet workspace and working headphones or a headset
  • Consistently available to work 10 or more hours per week
  • Able to begin onboarding and taking on tasks immediately within the next 1 to 2 days

How this role compares

This role’s top listed rate is $4/hr, below the $30/hr median for the 125 open Multimodal roles that list hourly pay (8 in 10 top out between $9 and $100/hr).

Across Terac’s 24 roles with hourly pay, the median is $90/hr, so this one is below it.

153 Multimodal roles are open on this board, from 7 platforms:

Terac’s listings were last checked on Nov 22, 2026.

Apply on Terac — this is a referral link; we may receive a reward if you’re hired, at no cost to you.

Listed on Terac. Details can change, so confirm on Terac before applying.

Similar roles

See all Multimodal jobs

Related AI trainer jobs