Duplicate from Ash11/ntina-voice-dataset-v0

Co-authored-by: MUHIGIRI Ashuza <Ash11@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1 @@
+*.wav filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,125 @@
+---
+language:
+  - sw
+  - yo
+  - en
+pretty_name: Ntina Voice Dataset V0
+task_categories:
+  - automatic-speech-recognition
+tags:
+  - code-switching
+  - swahili
+  - yoruba
+  - education
+  - reading-assessment
+license: other
+---
+
+# Ntina Voice Dataset V0
+
+V0 contains read-aloud and prompt-based recordings from one adult Swahili
+speaker from the Democratic Republic of the Congo, two adult Yoruba speakers,
+and two Yoruba child participants aged 10–15. It is intended for development
+of the Ntina TaRL reading-assessment benchmark.
+
+## Development status
+
+This is a public development V0 dataset. Contributor consent is confirmed for
+all speakers, and guardian consent is confirmed for both child participants.
+Questions and corrections can be sent to `ashuza1411@gmail.com`.
+
+Hugging Face: [Ash11/ntina-voice-dataset-v0](https://huggingface.co/datasets/Ash11/ntina-voice-dataset-v0)
+
+## Contents
+
+- 30 isolated syllables
+- 40 isolated words
+- 2 paragraphs
+- 2 stories
+- 4 comprehension-question recordings
+- 10 Swahili scripted Set B prompt recordings
+- 20 code-switched Set C scaffolding recordings
+- 86 Yoruba–English Set C scaffolding recordings
+- 10 Yoruba–English scripted Set B prompt recordings
+- 58 Yoruba child Set A recordings: 29 syllables, 22 words, 2 paragraphs,
+  1 story, and 4 comprehension questions
+- 262 WAV files total
+
+Ten Yoruba child Set B source responses do not have verbatim transcripts and
+are deliberately excluded from V0. They are retained outside this package for
+a possible later dataset version.
+
+All distributed clips are mono, 16 kHz, signed 16-bit PCM WAV. The original
+MP4 recordings remain outside this prepared dataset directory.
+
+## Structure
+
+- `audio/` — one WAV file per item
+- `metadata.csv` — transcript, item, speaker, duration, format, source, and
+  provenance fields
+- `speakers.csv` — pseudonymous speaker metadata
+
+Contributors are represented only by the pseudonymous IDs `spk-sw-001`,
+`spk-yo-001`, `spk-yo-002`, `spk-yo-003`, and `spk-yo-004`.
+The final two IDs represent different child participants: `spk-yo-003`
+recorded the first Set A folder and `spk-yo-004` recorded the second Set A
+folder. Filenames contain no real name.
+
+For Yoruba Set C, the `yoruba_Recording C2 - 45` source folder maps to the
+first 44 scaffold acts. Source recordings C46 through C87 map to the
+remaining 42 acts. C88 is not present and is not expected in V0.
+
+## Processing
+
+The syllable, word, and comprehension recordings were split at the midpoint
+of detected silence intervals. Detection used `-35 dB`, with a minimum gap of
+0.6 seconds for syllables and 1.0 second for words and comprehension items.
+Paragraph and story recordings were converted as complete clips. All outputs
+were resampled to mono 16 kHz PCM.
+
+The preparation process is reproducible with:
+
+```bash
+node tools/prepare-seta-v0.js
+node tools/prepare-yoruba-setc-v0.js
+node tools/prepare-yoruba-child-v0.js
+node tools/finalize-dataset-v0.js
+```
+
+## Validation and publication
+
+The prepared directory contains 262 metadata rows matching exactly 262 audio
+files. There are no missing or extra clips or blank transcripts. All 58 child
+clips are mono, 16 kHz, signed 16-bit PCM.
+
+The public Hugging Face repository is
+[Ash11/ntina-voice-dataset-v0](https://huggingface.co/datasets/Ash11/ntina-voice-dataset-v0).
+Audio is ignored by the
+application Git repository to prevent it from being pushed to GitHub; WAV
+files in the separate dataset repository are configured for Git LFS.
+
+## Limitations
+
+- One speaker is not representative of Congolese Swahili diversity.
+- Four Yoruba speakers are not representative of Yoruba dialect diversity.
+- The Yoruba Set C reference text and recording alignment were verified by a
+  contributor for V0.
+- “Congolese Swahili” is not a sufficiently precise dialect label; the
+  speaker's region must be recorded before dialect claims are made.
+- Audio boundaries and alignment were confirmed by the contributor for V0.
+- Set B contains scripted readings of the prompt text, not spontaneous
+  free-response answers for the adult contributors. Untranscribed child Set B
+  responses are excluded from V0 and reserved for a later version.
+- The child Set A session is partial. Missing items are `yo-ls-020`,
+  `yo-w-001`–`yo-w-010`, `yo-w-012`–`yo-w-018`, `yo-w-020`, and `yo-s-002`.
+- This V0 contains monolingual Set A reading material, not spontaneous
+  code-switched speech.
+
+## Consent, licensing, and access
+
+Contributor consent is recorded for every speaker, and guardian consent is
+recorded for both child participants. Confirmation was entered on 2026-08-15;
+the original document-signing date was not supplied. V0 uses custom
+development terms (`other` in Hugging Face metadata). Contact
+`ashuza1411@gmail.com` with reuse questions. Source texts retain their
+original attribution and license requirements.

audio/sw-en_setA_SPEAKER_sw-ls-001.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fde2108218fed6fc53327939c089925ae1c46bfce6159ba2d53b525ace4d791c
+size 61540

audio/sw-en_setA_SPEAKER_sw-ls-002.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27e78968db79409362992a6883977f032a68d15a527907643b7f6e8c5f2deb13
+size 48638

audio/sw-en_setA_SPEAKER_sw-ls-003.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42782223dfa1c8533535349bde2781b7a5000b8de1822da27068b30fc915f19c
+size 45800

audio/sw-en_setA_SPEAKER_sw-ls-004.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6a9f1ef38e54dfc63e221c287d6cc5fbc65d32341b75d0cf7bad966d1fc7a39
+size 45380

audio/sw-en_setA_SPEAKER_sw-ls-005.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce117421e27475dce671ebe060949ed33290b62d67c52131f0d9cd314103cc26
+size 45630

audio/sw-en_setA_SPEAKER_sw-ls-006.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbfe3bef0d710fb3e31c2062df77138f87c7aa88aaf57925bb483cdce09242e7
+size 42364

audio/sw-en_setA_SPEAKER_sw-ls-007.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e702b5f92f6823b6e21654762d858c92aa00fe43dc571052210b4f479e8a5d3
+size 52218

audio/sw-en_setA_SPEAKER_sw-ls-008.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4078bbe8f3d5c33770694b80e96ab28b15aa0f617c92aa6eda5376d0a384c345
+size 46868

audio/sw-en_setA_SPEAKER_sw-ls-009.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ea7ec9cffc57875c7c8db7c55ea0da2250b94a47e3b4d136da64915f7bba195
+size 47136

audio/sw-en_setA_SPEAKER_sw-ls-010.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d455b611f4dbf7dc30af16e0ddefef442762777fce77f0e076b123f27a8f0406
+size 54072

audio/sw-en_setA_SPEAKER_sw-ls-011.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02d640ea72562ad75a707e9448cdfdb1f7a1dbb866cce0634f976f38f27a65ce
+size 38026

audio/sw-en_setA_SPEAKER_sw-ls-012.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b06a5cee63017abaea5e1560dac81978e46596a1e6a98c9ffc32528a21b48a6
+size 51250

audio/sw-en_setA_SPEAKER_sw-ls-013.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:671fa23abb8ee794637ca6c6faecab20ffc1a259d30371585c23b2d7af048805
+size 52514

audio/sw-en_setA_SPEAKER_sw-ls-014.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63dbf023e7c84b207b565572edb13337a051dca22ee41867f44f7ecbe740435a
+size 47858

audio/sw-en_setA_SPEAKER_sw-ls-015.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2752052a9265b1251ad0b934188edd91852261ce915d43e3c2b1314e1a73f3a
+size 48910

audio/sw-en_setA_SPEAKER_sw-ls-016.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:013400fd3b5d5801719e9ece400740838e8f80b0e29ec34994ae9f657752497c
+size 51762

audio/sw-en_setA_SPEAKER_sw-ls-017.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:173839434c51fff2076ad34cf6c98825ae8dcecc18cfdcf34a1ff29c52a45e6c
+size 51174

audio/sw-en_setA_SPEAKER_sw-ls-018.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:182bebad57fa9511aa58c39de8b7af82ba81aec1bd99452c361ace85a0949c1d
+size 49028

audio/sw-en_setA_SPEAKER_sw-ls-019.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bb76610d178b3f4cf11f5dfc13099eb5ded3b758671a32baf86231d8f3a7fd1
+size 48280

audio/sw-en_setA_SPEAKER_sw-ls-020.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7032099e9fafe603ff551e2de7d975ce6a764f13a426266c8561662a445596a2
+size 47346

audio/sw-en_setA_SPEAKER_sw-ls-021.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6bc589a51e1750c747cd4e4c89e701c3d339c791995499d0eab90e7b7063b4fb
+size 51304

audio/sw-en_setA_SPEAKER_sw-ls-022.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe6274057ff5e6d16deb969774252bb430564021902be2761cf162e62fe7bbf5
+size 52228

audio/sw-en_setA_SPEAKER_sw-ls-023.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20f1bb039dd45245dee08d64059e41b229cf41f4cee2ee101a0b5da3d13786a5
+size 50552

audio/sw-en_setA_SPEAKER_sw-ls-024.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e368f342750a05a520d582c5e89907dcaed98a9249cb15b895f3f950dc514f0
+size 50914

audio/sw-en_setA_SPEAKER_sw-ls-025.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dba3890570d62447b44243b8a43ca118e53354cf9d139ae2e9cb9a6ce33eb506
+size 49098

audio/sw-en_setA_SPEAKER_sw-ls-026.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fd20986325be71e37301ef868229348a68fc280c0907778e6895853e2a0edad
+size 52118

audio/sw-en_setA_SPEAKER_sw-ls-027.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa09d66cbbbd035e14726083f528836a8262803e49d4a0ddbfdb89728fab98f3
+size 63996

audio/sw-en_setA_SPEAKER_sw-ls-028.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8802517d2debeee660c03bdd05004378cf2d36f983527985f5c405561f698ff
+size 45802

audio/sw-en_setA_SPEAKER_sw-ls-029.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f0ff5cc0941865e9504a22742cf17c283330533e5f7026887c923c7f934438d
+size 51076

audio/sw-en_setA_SPEAKER_sw-ls-030.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c2bb5dd68e3188486c91bae3923b15ac7597ee3368e36acb829843e382cfb36
+size 58100

audio/sw-en_setA_SPEAKER_sw-p-001.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4f74712b9d1eb5464bfa55327289d6580dc37db3caf26c1d58812d6ab9a935f
+size 573518

audio/sw-en_setA_SPEAKER_sw-p-002.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53f7eea027d78e8e1f37b054e0d0ef97cf40a534f82997d04fa296e73899a879
+size 610382

audio/sw-en_setA_SPEAKER_sw-s-001-q1.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f1891d5fd5a45e29ce68991fef2cf27256af44fb07b0d157638a1feabb9d363
+size 243918

audio/sw-en_setA_SPEAKER_sw-s-001-q2.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75a3ad78c73aa525ce30c6bf2d4ab61b58020332e8be4a4b19f0e5de13854a34
+size 292224

audio/sw-en_setA_SPEAKER_sw-s-001.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb2eedd5c1b06d77639272de62997ebc4c244f8bbd9409189b139c51c25287ee
+size 2321144

audio/sw-en_setA_SPEAKER_sw-s-002-q1.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c45a961840a598cda9b1b9ff0119602875e3bbd55e069de96fc999143d30a61d
+size 240400

audio/sw-en_setA_SPEAKER_sw-s-002-q2.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acd6de41ee1647f10263db87021e2bc5f044a291a3a6e74e71b653526da3b4ed
+size 286602

audio/sw-en_setA_SPEAKER_sw-s-002.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61b06a3fb2ed9309bf6b4426836978751367677b2294de0feb068d40d27c93df
+size 2502734

audio/sw-en_setA_SPEAKER_sw-w-001.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70dacb92825dd05ec59ad5364fc5925948358b925b88d3c398300e92f39cab04
+size 59778

audio/sw-en_setA_SPEAKER_sw-w-002.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4416b8598a4251e3cec8805066e50d44e7e6dc4eee399708251a803d02c23f7
+size 56882

audio/sw-en_setA_SPEAKER_sw-w-003.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4e0a8eb5210695e52f74e8675ae9145c9c556a6dee3c330e2437ead5c9e2e31
+size 57514

audio/sw-en_setA_SPEAKER_sw-w-004.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3813929f0b4bb5a3b49de43d9d6df3cf13e694ffae0ea1e2f4b2636699ba5fa6
+size 51324

audio/sw-en_setA_SPEAKER_sw-w-005.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89f7154bb8dd56ce66f981cd81acc3c9ab54f406e6a2c7219d7e3df0d4de364b
+size 52514

audio/sw-en_setA_SPEAKER_sw-w-006.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5083a2164a19bf47540490e687b0a8bc5afafae453f8816b518bf26d93b73ff0
+size 52332

audio/sw-en_setA_SPEAKER_sw-w-007.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:feab5da3a0b8c8ab23dac656fb18ce3763caac41a6726c1a3875f2531c0056a7
+size 56374

audio/sw-en_setA_SPEAKER_sw-w-008.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07a7d4b214b862510208ca0677b5b388781ad845a8b14c0baa8f8f2c0daa8131
+size 53286

audio/sw-en_setA_SPEAKER_sw-w-009.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88b0757f611bcec59ff4698aa5c853093630074c47b72bacfbac97eb927532fb
+size 54170

audio/sw-en_setA_SPEAKER_sw-w-010.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66fd2b0514228b6bfa4c905390b0c52c7e714725e12b4d46b1fd09ac2415aa23
+size 52610