Chinese pronunciation recordings

Speaker: Yue Tan.
Source: Audio-cmn / Shtooka / SWAC Recorder.
Repository: https://github.com/hugolpz/audio-cmn
Project management, compression and file renaming: Hugo Lopez (PLIDAM, INALCO).
Recording software and technical support: Nicolas Vion.

The Audio-cmn repository identifies these recordings as CC-by-sa.
Creative Commons Attribution-ShareAlike: https://creativecommons.org/licenses/by-sa/
See the preserved upstream README at AUDIO-UPSTREAM-README.txt for its license notice.

This app redistributes original unchanged Audio-cmn MP3 recordings with ASCII identifiers.
Selected vocabulary recordings use 64k/hsk (Yue Tan), upgraded from 24k-abr/hsk.
Other legacy recordings remain 24k-abr/hsk. Both retain their upstream CC-by-sa license.
The 64k selection, source URL, checksums and speaker credit are in data/human-audio-provenance.json.
Recordings are packed byte-exactly into small binary bundles; original source bytes are preserved.
data/audio-map.json maps Chinese filenames to their identifiers.
data/audio-pack-index.json lists the exact byte offsets of each unchanged MP3 in audio-packs.
Recordings are provided free of charge. They are from an older vocabulary collection;
HSK level labels in this app come from the separate 2025 syllabus.
Sequential playback of multiple recordings is explicitly identified as separate words/characters,
and is not a natural sentence recording. These files retain their source license.