Chorti Speech to Text: Revitalizing a Mayan Language Through AI

Chorti (Ch'orti') is a Mayan language spoken primarily in the eastern highlands of Guatemala, with a small number of speakers in Honduras. It belongs to the Cholan branch of the Mayan family, which also includes Chol and Chontal. Despite its rich history—descended from the language of the Classic Maya civilization—Chorti is now endangered, with only around 30,000 native speakers. Most Chorti speakers are bilingual in Spanish, and younger generations are shifting away from the language. Accurate speech-to-text technology offers a powerful tool for language preservation and daily use.

Why Accurate Chorti Speech-to-Text Matters

For the Chorti community, transcription and subtitling are not just conveniences—they are acts of cultural survival. Oral histories, traditional songs, and agricultural instructions are passed down verbally. By converting these into written text, communities can create archives that are searchable and shareable. Subtitled videos in Chorti on social media can reach younger speakers and the diaspora, reinforcing the language's presence in digital spaces. Accessible transcription also supports bilingual education, where Chorti is used alongside Spanish in classrooms.

Specific Transcription Challenges

Chorti presents several hurdles for automatic speech recognition:

• Ejective consonants: Sounds like t', k', ch', and p' require careful acoustic modeling. Mishearing them can change the meaning entirely.
• Glottal stop: Marked with an apostrophe, it is a consonant that must be detected.
• Vowel length: Minimal pairs exist, e.g., 'tam' (to cover) vs. 'taam' (to be deep).
• Dialectal variation: Differences in pronunciation between Jocotán, Camotán, and La Unión can affect accuracy.
• Code-switching: Speakers often mix Chorti and Spanish, requiring models to handle both languages.

Speechyou's models are built to handle these features. They are trained on a corpus that includes recordings from various dialects and are continuously updated based on user feedback.

Use Cases for Chorti Transcription

• Preserving oral literature: Transcribe stories from elders before they are lost.
• Church services: Many congregations hold mass in Chorti; transcripts help with liturgy preparation.
• Community radio: Stations like Radio Chorti can subtitle their programs for wider reach.
• Linguistic research: Phonetic analysis of Chorti becomes easier with accurate text.
• YouTube subtitles: Creators can add Chorti subtitles to videos, increasing engagement.
• Medical and legal translation: Ensure Chorti speakers have access to information in their language.

How Speechyou Helps

Speechyou provides a dedicated Chorti speech-to-text engine that is accessible via web browser. No technical expertise is required. Simply upload your audio, choose Chorti, and receive a transcript that can be exported as SRT, VTT, or plain text. The tool is cost-effective, with unlimited transcription included in the Solo plan—ideal for non-profit organizations or community groups. By supporting low-resource languages like Chorti, Speechyou contributes to linguistic diversity and empowers communities to maintain their heritage in the digital world.