Find the spoken moment
Search indexed video captions for a word or phrase and jump to matching moments in YouTube videos.
Hear the conversation around it
Read nearby captions and open the transcript to hear how pronunciation, rhythm, and meaning fit the conversation.
Control how you listen
Adjust playback speed, replay the caption window, and repeat short sections for listening and shadowing.
Record and compare
Record a take in your browser and play it back alongside the example. Microphone permission and browser support are required. Comparison is by ear; Tubonics does not provide automated pronunciation scoring.
Choose your context
Use topic, video, channel, sentence-type, and supported word-class filters. Speaker options infer hints from titles and channels and may retain unknown speakers.
Discover and practice
Explore language collections, vocabulary, topics, and indexed videos. Guided drills currently cover Japanese and English; replay, repetition, and recording work with clips across spoken languages.
Explore available guidance
Read available meanings, readings, and phonetic notes. Japanese lookups include Hepburn, mora timing, pitch notes, English glosses, and gojūon browsing. Guidance varies by language and word.
Bring Tubonics into your site
Use the documented Tubonics embed and JavaScript interface to include pronunciation search and listening in another site.
What determines an available example?
Tubonics searches its indexed caption corpus first. Captions can be uploaded or autogenerated, and may contain transcription or timing errors. A caption match identifies a video moment; it does not certify the speaker’s pronunciation or guarantee that YouTube still permits playback. Region and speaker options use metadata hints where enabled. Detailed guidance and drills vary by language.
See language coverage and regional options, search filter syntax, and developer documentation.