thelarryproject/scotus-index-vocoder

thelarryproject SCOTUS Vocoder Index

An index over thelarryproject/scotus (Supreme Court oral argument segments): the clips that pass the audio quality and transcript filters used for vocoder training, as row numbers and ids only. Each row points at its clip in the source dataset by row number.

182,282 of the source's 384,222 clips remain:

split	clips	hours	speakers
train	182,055	761.4	918
validation	227	1.0	7

The splits are speaker-disjoint.

Fields

• row: the clip's row number in the source's train split; both splits point into it
• id: the source's clip id (docket:start_s), e.g. 08-1438:0.2; check it against the source row to confirm the match

Construction

Rows are numbered first. Removed, with the stricter thresholds used for synthesis data:

• clips shorter than 0.1 s or longer than 25 s
• clips with ctc_mismatch above 0.2
• clips failing an audio quality check: overall level below -40 dBFS, speech in under 50% of frames, signal-to-noise ratio below 15 dB, or more than 0.01% of samples clipped

Every remaining clip from eight advocates goes to validation: Glen Gifford, William Brockman, Susan Smith, Nathaniel Zelinsky, Ann Adams, Krissy Nobile, Meir Feder, and Margie Phelps. Seven of them have clips that pass the filters. Then every column but row and id is dropped. See provenance.json for the producing commit and timestamp.

Known corrections

N/a; corrections made while aligning the transcripts are listed in the source dataset's card

Known gaps

• Only row and id: read audio, transcripts, and every other column from the source dataset by row; a later revision of the source may reorder rows

The audio and transcripts are Public Domain, but this redistribution is CC-BY-4.0 under thelarryproject.

Preprocessing pipeline code is at https://github.com/Mekadrom/larry, and this dataset is fully reproducible with a single command in that repo:

uv run preprocessing -f configs/preprocessing/index/voice/asr/thelarryproject_scotus-index-asr.yaml -f configs/preprocessing/index/voice/vocoder/thelarryproject_scotus-index-vocoder.yaml --output_dir ~/.cache/larry/datasets/index/voice/vocoder/thelarryproject_scotus-index-vocoder --clean