jsonl764 B

ASR Digit Utterances Dataset (JSONL)

JSON Lines ASR training/eval set for the Wave B synthetic digit utterances — each row points at a clean WAV and carries the expected transcript.

ASR testing ML training data NLP datasets

Preview — first 9 linesjsonl

{"id": "u001", "audio": "digits-0123-clean.wav", "text": "zero one two three", "duration_ms": 900}
{"id": "u002", "audio": "digits-4567-clean.wav", "text": "four five six seven", "duration_ms": 900}
{"id": "u003", "audio": "digits-89-clean.wav", "text": "eight nine", "duration_ms": 500}
{"id": "u004", "audio": "pin-1357-clean.wav", "text": "one three five seven", "duration_ms": 900}
{"id": "u005", "audio": "seq-24680-clean.wav", "text": "two four six eight zero", "duration_ms": 1100}
{"id": "u006", "audio": "seq-987654-clean.wav", "text": "nine eight seven six five four", "duration_ms": 1300}
{"id": "u007", "audio": "yes-1-clean.wav", "text": "one", "duration_ms": 300}
{"id": "u008", "audio": "no-0-clean.wav", "text": "zero", "duration_ms": 300}

Specifications

Records: 8
Task: ASR
Schema: id, audio, text, duration_ms
Note: pairs with audio/asr/*-clean.wav fixtures

What is a .jsonl file?

JSONL (JSON Lines) is a text format where each line is a complete, independent JSON value, allowing records to be streamed and appended without parsing the whole file. It is not itself a JSON array and each line must stand alone. It is common in logging, machine learning datasets, and data pipelines.

How to use this file

Use an example JSONL to test line-by-line streaming parsers, append-and-resume ingestion, and batch pipelines that process one record per line.

How to use this file for testing

“ASR Digit Utterances Dataset (JSONL)” is a deterministic Novus Examples fixture for ASR testing, ML training data, NLP datasets. Short synthetic digit/tone utterances with transcript JSON and clean↔noise pairs — for testing ASR loaders, WER harnesses, and audio preprocessing.

Documented properties for this file: 8 records · schema: id, audio, text, duration_ms · task: ASR. Compare results against paired or grouped companions on this page when present (clean↔damaged, searchable↔scanned, or format twins) so scores stay reproducible across runs.

Download the file once, keep the path stable in CI or local scripts, and treat the spec table as the contract: dimensions, seeds, field lists, and roles are intentional. Corrupt or invalid samples are labelled as such — expect parsers to fail loudly rather than silently accept them.

AI/ML fixtures are fully synthetic with documented schemas — no real people or data. Test data loaders, tokenizers, annotation converters, embedding/vector stores, or eval-metric parsers against the known structure and fixed seeds.

Run ASR on the clip and score the transcript against the documented ground-truth script; clips are short, synthetic, and read from a fixed word or digit sequence.

Code examples

import json

with open("asr-digit-utterances.jsonl") as f:
    rows = [json.loads(line) for line in f]
print(len(rows), rows[0])

Generated by generation/ai_wave_b.py. Free for any use, no attribution required — license.