Bilingual Mix

It automatically aligns the Chinese and English versions of the same audio (dubbed cartoons, audiobooks, textbook recordings) sentence by sentence and produces alternating practice audio: one line Chinese, then the same line in English. This is a common format for comprehensible-input language learning.

Audio is sent to Groq for sentence timestamps; transcript text is sent to Cloudflare Workers AI for cross-language semantic pairing. Raw audio is not stored, and the final file is cut and joined in your browser.

MP3 files up to 50 MB (about 45 minutes) each. The Chinese and English versions must contain the same content in the same order.

The output is a standard MP3 (192 kbps) that works on phones, computers, cars, and dedicated listening devices.

Purely instrumental and sung passages are skipped automatically — they cannot be aligned reliably sentence by sentence. Skipping them keeps every line in the result a clean Chinese-English pair.