Khetrani Speech to Text: Preserving a Minority Language Through AI

Khetrani (کھیترانی) is a little-known Indo-Aryan language spoken in the rugged hills of Balochistan, Pakistan. With an estimated 50,000 to 100,000 speakers, it is a minority language even within its own region, surrounded by Balochi, Saraiki, and Pashto. Despite its small size, Khetrani has a vibrant oral culture: folk tales passed down through generations, poetic traditions, and a unique identity that its speakers fiercely protect. However, the language lacks a standardized written form and has very few digital resources, making it vulnerable to language shift.

Why Accurate Speech to Text Matters for Khetrani

In an age where digital content dominates, Khetrani speakers are often left behind. There is no Google Translate for Khetrani, no major media outlets broadcasting in the language, and few educational materials. Accurate speech-to-text transcription can change that. By converting spoken Khetrani into written text, we can:

• Create subtitles for videos in Khetrani, making them accessible to deaf community members and diaspora.
• Document oral histories and folklore for future generations.
• Enable researchers to study the language's grammar and phonology.
• Support language revitalization by providing written models for learners.

Without transcription tools, these efforts are manual, slow, and expensive. Speechyou offers an AI-powered solution that works directly with Khetrani audio, generating text in the Arabic script with high accuracy.

Specific Challenges in Transcribing Khetrani

Khetrani poses unique challenges for automatic speech recognition (ASR). Its phonology includes retroflex consonants (like /ʈ/ and /ɖ/), which are rare in many languages, and a complex system of aspirated and murmured stops. Vowel length is phonemic, meaning a short vs. long vowel can change a word's meaning. The Arabic script adds another layer: it is written right-to-left, uses diacritics for short vowels, and has contextual letter forms. Most ASR systems are trained on languages like English or Mandarin and fail to capture these features.

Dialectal variation further complicates matters. Khetrani spoken in Barkhan differs from that in Kohlu, with vocabulary and pronunciation shifts. Speechyou's model is trained on a diverse corpus of Khetrani speech from multiple regions, allowing it to adapt to these variations. The system also supports custom vocabulary, so users can add local terms or names.

Use Cases: From Podcasts to Preservation

Khetrani speakers are increasingly creating content on platforms like YouTube and Facebook, but without subtitles, their reach is limited. Speechyou enables automatic subtitle generation in SRT and VTT formats, which can be uploaded directly to video platforms. This is especially valuable for religious sermons, cultural events, and educational videos.

Researchers in linguistics and anthropology can use Speechyou to transcribe field recordings, saving hours of manual work. The timestamped output allows for easy alignment with audio, facilitating phonetic analysis. For community activists, transcribing oral histories is a step toward creating a written archive of Khetrani heritage.

How Speechyou Helps

Speechyou's technology is built for low-resource languages. Instead of requiring massive datasets, it uses transfer learning from related languages (like Saraiki and Punjabi) to bootstrap accuracy. The interface is simple: upload an audio file, select Khetrani, and receive a transcription within minutes. The output can be edited, exported as plain text, or converted into subtitles with proper RTL support.

For a language that has been largely invisible to technology, Speechyou provides a vital bridge. Whether you are a Khetrani speaker wanting to subtitle your videos, a researcher documenting folklore, or a teacher creating materials, Speechyou gives you the tools to work in your own language.

Conclusion

Khetrani may be a small language, but its speakers deserve the same digital tools as speakers of major languages. Speechyou's speech-to-text transcription and subtitle generation for Khetrani is a step toward linguistic equality. By preserving oral traditions and enabling new forms of communication, we help ensure that Khetrani thrives in the modern world.