STT — Validated Upstream Inputs
Collection
The 14 upstream repositories used by the completed STT training run. These are not final immutable training releases. • 14 items • Updated
Datasets:
Request access to Instinct speech data
Requests are reviewed manually. Approval is not automatic and may be revoked if the dataset terms are violated.
By submitting this request, you agree to use the dataset only for the stated purpose, follow its license and privacy terms, and not redistribute the data.
Log in or Sign Up to review the conditions and access this dataset content.
Forced-Aligned STT Dataset
Source dataset: instinct-org/cv_chunked
Aligned dataset: instinct-org/cv_chunked_nfa_aligned
Rows: 71097 successfully aligned rows
This dataset adds NeMo Forced Aligner metadata for STT training and timestamp quality control. Source rows that did not produce usable CTM alignments are excluded from the published data.
Added columns:
nfa_token_alignments: NeMo token/subword CTM spansnfa_word_alignments: word-level CTM spansnfa_segment_alignments: segment-level CTM spansnfa_character_alignments: character spans; true only when the model token CTM is character level, otherwise derived from word spansnfa_character_alignment_sourcenfa_true_character_alignmentnfa_alignment_modelnfa_alignment_statusnfa_alignment_error
NeMo Forced Aligner emits token, word, and segment CTM levels for this model family. With BPE/subword ASR models, token level is not the same as character level.
Model: nvidia/stt_uz_fastconformer_hybrid_large_pc
Data product classification
- Family:
cv - Purpose:
stt - Lifecycle state:
aligned - Schema:
stt-v1 - Access policy:
public-manual-v1(manual approval) - Catalog: instinct-org/speech-data-catalog
Exact parent revisions and processing recipes are being registered in the catalog. Until that lineage review is complete, do not infer processing order solely from the repository name.
- Downloads last month
- 20