What is Speech-to-Text (STT)?

Short Answer #

Speech-to-Text (STT) technology converts spoken language into written text. In language revitalization, it can help communities transcribe recordings, interviews, stories, lessons, and conversations more efficiently.

Expanded Answer #

Every language revitalization program eventually faces the same challenge. There are more recordings than there is time to transcribe. Many communities possess:

• Elder interviews
• Oral histories
• Language lessons
• Community recordings
• Storytelling sessions
• Cultural teachings
• Historical archives

These recordings contain valuable knowledge, but they are often difficult to search and use unless they are transcribed. Speech-to-Text technology helps address this challenge by converting audio into text. When functioning effectively, STT can help language teams:

• Reduce transcription workloads
• Improve searchability
• Create educational materials
• Build dictionaries
• Support language learning
• Prepare data for future technologies

However, STT performance depends heavily on the quality and quantity of available language data. Major world languages benefit from decades of technology development and enormous training datasets. Many Indigenous languages do not yet have access to the same resources. For this reason, Indigenous language STT development often requires significant community participation, recording collection, transcription work, and language corpus development. Language Cloud helps communities organize the resources needed to support this process.