Update dataset card (pld.md)
Browse files
README.md
CHANGED
|
@@ -10,49 +10,119 @@ language:
|
|
| 10 |
- pam
|
| 11 |
- tsg
|
| 12 |
- war
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
task_categories:
|
| 14 |
- automatic-speech-recognition
|
| 15 |
- text-to-speech
|
|
|
|
|
|
|
| 16 |
tags:
|
| 17 |
- philippines
|
| 18 |
- philippine-languages
|
| 19 |
- low-resource
|
| 20 |
-
- speech
|
| 21 |
- multilingual
|
| 22 |
-
|
| 23 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
---
|
| 25 |
|
| 26 |
# Philippine Language Dataset (PLD)
|
| 27 |
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
|
| 37 |
-
|
|
|
|
|
|
|
|
|
|
| 38 |
|
| 39 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
|
| 41 |
Rows where `speech_type == "spontaneous"` do **not** carry a transcript. The
|
| 42 |
-
session logs store the *elicitation question* that was put to the speaker
|
| 43 |
-
|
| 44 |
their free-speech answer. The same question text repeats verbatim across
|
| 45 |
-
different speakers.
|
| 46 |
-
|
|
|
|
|
|
|
|
|
|
| 47 |
|
| 48 |
```python
|
| 49 |
ds = ds.filter(lambda x: not x["text_is_prompt"])
|
| 50 |
```
|
| 51 |
|
| 52 |
-
|
| 53 |
-
useful for self-supervised pretraining, speaker modeling, or re-transcription.
|
| 54 |
|
| 55 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
|
| 57 |
| Code | Language | Utterances | Hours |
|
| 58 |
|---|---|---|---|
|
|
@@ -67,53 +137,34 @@ useful for self-supervised pretraining, speaker modeling, or re-transcription.
|
|
| 67 |
| `pag` | Pangasinan | 5,566 | 8.7 |
|
| 68 |
| `tsg` | Tausug | 4,495 | 5.9 |
|
| 69 |
|
| 70 |
-
English word and sentence lists (`EngW.txt`, `EngSen.txt`) were read by the
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
|
| 75 |
-
## Speech types
|
| 76 |
|
| 77 |
-
| Type | Utterances | Hours | Mean
|
| 78 |
-
|---|---|---|---|
|
| 79 |
-
| `read` | 158,121 | 302.9 | 6.
|
| 80 |
-
| `isolated` | 164,447 | 107.3 | 2.
|
| 81 |
-
| `spontaneous` | 2,586 | 32.2 | 44.
|
| 82 |
-
| `digits` | 9,114 | 5.9 | 2.
|
| 83 |
-
|
| 84 |
-
- `read` — full prompted sentences (news, medical, literature, essays); the
|
| 85 |
-
material best suited to TTS
|
| 86 |
-
- `isolated` — single words and short phrases from word lists
|
| 87 |
-
- `digits` — spoken digit strings
|
| 88 |
-
- `spontaneous` — free speech; **prompt-only text**, see the warning above
|
| 89 |
|
| 90 |
-
##
|
| 91 |
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
ds = ds.filter(lambda x: not x["text_is_prompt"] and 0.3 <= x["duration"] <= 30.0)
|
| 97 |
-
ds = ds.cast_column("audio", Audio(sampling_rate=16000))
|
| 98 |
-
```
|
| 99 |
-
|
| 100 |
-
**TTS finetuning (read speech only):**
|
| 101 |
-
```python
|
| 102 |
-
ds = ds.filter(lambda x: x["speech_type"] == "read"
|
| 103 |
-
and 1.0 <= x["duration"] <= 15.0
|
| 104 |
-
and x["num_words"] >= 3)
|
| 105 |
-
```
|
| 106 |
|
| 107 |
-
|
| 108 |
-
```python
|
| 109 |
-
ds = ds.filter(lambda x: x["language"] == "bcl") # Bikol
|
| 110 |
-
```
|
| 111 |
|
| 112 |
## Schema
|
| 113 |
|
| 114 |
| Field | Type | Description |
|
| 115 |
|---|---|---|
|
| 116 |
-
| `audio` | `Audio(16000)` |
|
| 117 |
| `sentence` | `str` | Prompt text read by the speaker (see `text_is_prompt`) |
|
| 118 |
| `duration` | `float` | Seconds |
|
| 119 |
| `num_words` | `int` | Whitespace word count |
|
|
@@ -128,27 +179,106 @@ ds = ds.filter(lambda x: x["language"] == "bcl") # Bikol
|
|
| 128 |
| `gender` | `str` | `male` / `female` / `unknown` |
|
| 129 |
| `age` | `int` | Speaker age, `-1` when unrecorded |
|
| 130 |
| `speaker_dialect` | `str` | Self-reported dialect |
|
| 131 |
-
| `mother_dialect` / `father_dialect` | `str` | Parents' dialects |
|
| 132 |
| `profession` | `str` | Self-reported profession |
|
| 133 |
| `session_id` | `str` | Recording session identifier |
|
| 134 |
| `session_environment` | `str` | Recording environment note |
|
| 135 |
| `source_file` | `str` | Original WAV stem |
|
| 136 |
|
| 137 |
-
|
|
|
|
|
|
|
| 138 |
|
| 139 |
-
|
| 140 |
-
|
| 141 |
-
|
| 142 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 143 |
|
| 144 |
-
|
| 145 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 146 |
|
| 147 |
-
##
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 148 |
|
| 149 |
-
|
| 150 |
-
the original corpus authors when using this data.
|
| 151 |
|
| 152 |
-
|
|
|
|
| 153 |
|
| 154 |
-
|
|
|
|
|
|
|
|
|
| 10 |
- pam
|
| 11 |
- tsg
|
| 12 |
- war
|
| 13 |
+
license: other
|
| 14 |
+
license_name: up-dsp-research
|
| 15 |
+
pretty_name: Philippine Language Dataset (PLD)
|
| 16 |
+
size_categories:
|
| 17 |
+
- 100K<n<1M
|
| 18 |
task_categories:
|
| 19 |
- automatic-speech-recognition
|
| 20 |
- text-to-speech
|
| 21 |
+
multilinguality:
|
| 22 |
+
- multilingual
|
| 23 |
tags:
|
| 24 |
- philippines
|
| 25 |
- philippine-languages
|
| 26 |
- low-resource
|
|
|
|
| 27 |
- multilingual
|
| 28 |
+
- speech
|
| 29 |
+
- bikol
|
| 30 |
+
- cebuano
|
| 31 |
+
- kapampangan
|
| 32 |
+
- ilocano
|
| 33 |
+
- hiligaynon
|
| 34 |
+
- waray
|
| 35 |
+
- pangasinan
|
| 36 |
+
- tausug
|
| 37 |
---
|
| 38 |
|
| 39 |
# Philippine Language Dataset (PLD)
|
| 40 |
|
| 41 |
+
**Ten Philippine languages, 980 speakers, 448 hours of prompted speech — one of the largest multilingual Philippine speech collections available as Parquet.**
|
| 42 |
+
|
| 43 |
+
<div align="center">
|
| 44 |
+
|
| 45 |
+
**334,268 utterances · 448.2 hours · 980 speakers · 10 languages · 16kHz mono**
|
| 46 |
+
|
| 47 |
+
[](https://github.com/sapinsapin/halohalo)
|
| 48 |
+
|
| 49 |
+
</div>
|
| 50 |
|
| 51 |
+
Collected by the **University of the Philippines Diliman Digital Signal
|
| 52 |
+
Processing Laboratory**. Every row is one prompted recording: the corpus ships
|
| 53 |
+
pre-segmented WAVs with the prompt text stored inline in each session log, so
|
| 54 |
+
no forced alignment or segmentation was applied here.
|
| 55 |
|
| 56 |
+
Most Philippine language speech data stops at Tagalog. This one covers Bikol,
|
| 57 |
+
Cebuano, Kapampangan, Hiligaynon, Ilocano, Waray, Pangasinan and Tausug at
|
| 58 |
+
scale — languages with tens of millions of speakers and almost no public ASR/TTS
|
| 59 |
+
data.
|
| 60 |
|
| 61 |
+
> **Read [`text_is_prompt`](#-read-this-before-training) before you train.** 2,586 rows
|
| 62 |
+
> carry an elicitation *question* instead of a transcript, and silently training
|
| 63 |
+
> on them will poison your model.
|
| 64 |
+
|
| 65 |
+
---
|
| 66 |
+
|
| 67 |
+
## 30-second quickstart
|
| 68 |
+
|
| 69 |
+
```python
|
| 70 |
+
from datasets import load_dataset
|
| 71 |
+
|
| 72 |
+
ds = load_dataset("sapinsapin/pld", split="train", streaming=True)
|
| 73 |
+
row = next(iter(ds))
|
| 74 |
+
|
| 75 |
+
print(row["language_name"], "|", row["sentence"])
|
| 76 |
+
print(row["speech_type"], row["duration"], "s")
|
| 77 |
+
```
|
| 78 |
+
|
| 79 |
+
The training-ready filter, in full:
|
| 80 |
+
|
| 81 |
+
```python
|
| 82 |
+
ds = load_dataset("sapinsapin/pld", split="train")
|
| 83 |
+
ds = ds.filter(lambda x: not x["text_is_prompt"] and 0.3 <= x["duration"] <= 30.0)
|
| 84 |
+
```
|
| 85 |
+
|
| 86 |
+
One language at a time:
|
| 87 |
+
|
| 88 |
+
```python
|
| 89 |
+
bikol = ds.filter(lambda x: x["language"] == "bcl")
|
| 90 |
+
```
|
| 91 |
+
|
| 92 |
+
---
|
| 93 |
+
|
| 94 |
+
## ⚠️ Read this before training
|
| 95 |
+
|
| 96 |
+
### `text_is_prompt` — 2,586 rows have no transcript
|
| 97 |
|
| 98 |
Rows where `speech_type == "spontaneous"` do **not** carry a transcript. The
|
| 99 |
+
session logs store the *elicitation question* that was put to the speaker —
|
| 100 |
+
e.g. *"Saen an dream destination mo?"* — while the audio is 20–90 seconds of
|
| 101 |
their free-speech answer. The same question text repeats verbatim across
|
| 102 |
+
different speakers.
|
| 103 |
+
|
| 104 |
+
They are kept because the audio is genuine spontaneous speech (32 hours of it,
|
| 105 |
+
valuable for pretraining, VAD, diarization, or re-transcription), but they are
|
| 106 |
+
poison for supervised `(audio, text)` training:
|
| 107 |
|
| 108 |
```python
|
| 109 |
ds = ds.filter(lambda x: not x["text_is_prompt"])
|
| 110 |
```
|
| 111 |
|
| 112 |
+
### Other things to know
|
|
|
|
| 113 |
|
| 114 |
+
| Gotcha | Detail | What to do |
|
| 115 |
+
|---|---|---|
|
| 116 |
+
| Prompts, not transcripts | Text is what the speaker was *asked* to read; no one verified they read it exactly | Treat as weakly-supervised; round-trip ASR to score |
|
| 117 |
+
| Speaker overlap | Random 90/10 split over utterances, so speakers appear in both splits | Re-split on `speaker_id` for speaker-disjoint eval |
|
| 118 |
+
| `eng` is not native English | English word/sentence lists read by Filipino L2 speakers | Use `corpus_language` to see which collection they came from |
|
| 119 |
+
| Half the corpus is single words | 164k `isolated` rows average 2.3 s | Filter on `speech_type` for sentence-level work |
|
| 120 |
+
|
| 121 |
+
---
|
| 122 |
+
|
| 123 |
+
## What's inside
|
| 124 |
+
|
| 125 |
+
### Languages
|
| 126 |
|
| 127 |
| Code | Language | Utterances | Hours |
|
| 128 |
|---|---|---|---|
|
|
|
|
| 137 |
| `pag` | Pangasinan | 5,566 | 8.7 |
|
| 138 |
| `tsg` | Tausug | 4,495 | 5.9 |
|
| 139 |
|
| 140 |
+
English word and sentence lists (`EngW.txt`, `EngSen.txt`) were read by the same
|
| 141 |
+
speakers. Those rows are labeled `language = "eng"`, while `corpus_language`
|
| 142 |
+
retains the Philippine language collection they came from, so per-language
|
| 143 |
+
filters stay clean either way.
|
| 144 |
|
| 145 |
+
### Speech types
|
| 146 |
|
| 147 |
+
| Type | Utterances | Hours | Mean | What it is |
|
| 148 |
+
|---|---|---|---|---|
|
| 149 |
+
| `read` | 158,121 | 302.9 | 6.9 s | Full prompted sentences — news, medical, literature, education, tourism. **Best material for TTS.** |
|
| 150 |
+
| `isolated` | 164,447 | 107.3 | 2.3 s | Single words and short phrases from word lists |
|
| 151 |
+
| `spontaneous` | 2,586 | 32.2 | 44.8 s | Free speech — **prompt-only text**, see the warning above |
|
| 152 |
+
| `digits` | 9,114 | 5.9 | 2.3 s | Spoken digit strings |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 153 |
|
| 154 |
+
### Splits
|
| 155 |
|
| 156 |
+
| Split | Rows |
|
| 157 |
+
|---|---|
|
| 158 |
+
| `train` | 300,842 |
|
| 159 |
+
| `test` | 33,426 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 160 |
|
| 161 |
+
---
|
|
|
|
|
|
|
|
|
|
| 162 |
|
| 163 |
## Schema
|
| 164 |
|
| 165 |
| Field | Type | Description |
|
| 166 |
|---|---|---|
|
| 167 |
+
| `audio` | `Audio(16000)` | 16 kHz mono, FLAC-compressed in storage |
|
| 168 |
| `sentence` | `str` | Prompt text read by the speaker (see `text_is_prompt`) |
|
| 169 |
| `duration` | `float` | Seconds |
|
| 170 |
| `num_words` | `int` | Whitespace word count |
|
|
|
|
| 179 |
| `gender` | `str` | `male` / `female` / `unknown` |
|
| 180 |
| `age` | `int` | Speaker age, `-1` when unrecorded |
|
| 181 |
| `speaker_dialect` | `str` | Self-reported dialect |
|
| 182 |
+
| `mother_dialect` / `father_dialect` | `str` | Parents' dialects — useful for contact/variation studies |
|
| 183 |
| `profession` | `str` | Self-reported profession |
|
| 184 |
| `session_id` | `str` | Recording session identifier |
|
| 185 |
| `session_environment` | `str` | Recording environment note |
|
| 186 |
| `source_file` | `str` | Original WAV stem |
|
| 187 |
|
| 188 |
+
The dialect fields are unusually rich for a speech corpus — speaker, mother and
|
| 189 |
+
father dialect are all recorded, which supports dialectometry and language-contact
|
| 190 |
+
work that most corpora can't.
|
| 191 |
|
| 192 |
+
---
|
| 193 |
+
|
| 194 |
+
## Models trained on this data
|
| 195 |
+
|
| 196 |
+
No reference finetune has been trained on PLD yet. The pipeline supports it —
|
| 197 |
+
add a `pld` branch to the dataset adapter and the existing trainers work
|
| 198 |
+
unchanged.
|
| 199 |
+
|
| 200 |
+
Meanwhile, the sibling Filipino corpus has working baselines you can copy the
|
| 201 |
+
recipe from:
|
| 202 |
+
|
| 203 |
+
| Model | Task | Trained on |
|
| 204 |
+
|---|---|---|
|
| 205 |
+
| [`sapinsapin/speecht5_tts-fsc`](https://huggingface.co/sapinsapin/speecht5_tts-fsc) | Text-to-speech | [`filipinospeechcorpus`](https://huggingface.co/datasets/sapinsapin/filipinospeechcorpus) |
|
| 206 |
+
| [`sapinsapin/whisper-small-fsc`](https://huggingface.co/sapinsapin/whisper-small-fsc) | Speech recognition | [`filipinospeechcorpus`](https://huggingface.co/datasets/sapinsapin/filipinospeechcorpus) |
|
| 207 |
+
|
| 208 |
+
**If you train something on PLD, tag this dataset in your model card** — it will
|
| 209 |
+
show up here automatically, and a Bikol or Cebuano ASR baseline would be the
|
| 210 |
+
first of its kind in public.
|
| 211 |
|
| 212 |
+
---
|
| 213 |
+
|
| 214 |
+
## How it was built
|
| 215 |
+
|
| 216 |
+
1. Walk every session directory; parse the per-session `.log` (speaker
|
| 217 |
+
demographics header, then one row per utterance: WAV name, prompt list,
|
| 218 |
+
prompt text).
|
| 219 |
+
2. Classify each utterance's `speech_type`. Explicit markers (`_Iso_`, `_Utt_`,
|
| 220 |
+
`Spontaneous`, digits) are used where present; the corpus uses at least five
|
| 221 |
+
naming conventions, so the ~55k rows with no marker are typed by the
|
| 222 |
+
**measured median word count** of their prompt list rather than by guessing
|
| 223 |
+
from the filename.
|
| 224 |
+
3. Repair double-encoded UTF-8 in transcripts (`hapúnan` → `hap��nan`) — 150 of
|
| 225 |
+
166 affected lines recover; the rest are left intact rather than risk a worse
|
| 226 |
+
string.
|
| 227 |
+
4. Resample to 16 kHz mono, encode FLAC, shard to Parquet, 90/10 random split.
|
| 228 |
+
|
| 229 |
+
1,943 rows (0.6%) reference WAVs that are not present in the archive and were
|
| 230 |
+
skipped.
|
| 231 |
+
|
| 232 |
+
Pipeline source: [`process_pld_parquet.py`](https://github.com/sapinsapin/halohalo/blob/main/process_pld_parquet.py)
|
| 233 |
+
· parser: [`halolib/pld.py`](https://github.com/sapinsapin/halohalo/blob/main/halolib/pld.py)
|
| 234 |
+
|
| 235 |
+
---
|
| 236 |
+
|
| 237 |
+
## Limitations
|
| 238 |
+
|
| 239 |
+
- **Prompted, not conversational.** Except for the 32 h spontaneous portion,
|
| 240 |
+
this is people reading from lists. Prosody and vocabulary reflect that.
|
| 241 |
+
- **Transcripts are unverified prompts.** Nobody checked that speakers read the
|
| 242 |
+
prompt exactly; expect a residual mismatch rate.
|
| 243 |
+
- **Coverage is uneven** — Bikol has 95.8 h, Tausug 5.9 h. Don't expect balanced
|
| 244 |
+
multilingual behaviour without resampling.
|
| 245 |
+
- **Recording conditions vary** by session and are only loosely described in
|
| 246 |
+
`session_environment`.
|
| 247 |
+
- **No held-out speaker split** is provided by default.
|
| 248 |
+
- Language codes follow ISO 639-3; `fil` and `tgl` distinctions in the wild are
|
| 249 |
+
inconsistent, so filter on both if you merge with other corpora.
|
| 250 |
+
|
| 251 |
+
---
|
| 252 |
|
| 253 |
+
## Related datasets
|
| 254 |
+
|
| 255 |
+
Part of the **halohalo** Philippine-language speech family:
|
| 256 |
+
|
| 257 |
+
| Dataset | What it covers | Scale |
|
| 258 |
+
|---|---|---|
|
| 259 |
+
| **pld** *(this one)* | 10 Philippine languages, prompted | 334k utterances · 448 h |
|
| 260 |
+
| [`sapinsapin/filipinospeechcorpus`](https://huggingface.co/datasets/sapinsapin/filipinospeechcorpus) | Filipino studio read + spontaneous | 305k segments · 65 h |
|
| 261 |
+
| [`sapinsapin/halo-livestream`](https://huggingface.co/datasets/sapinsapin/halo-livestream) | Taglish code-switched livestream speech | seed release |
|
| 262 |
+
|
| 263 |
+
---
|
| 264 |
+
|
| 265 |
+
## License, source and citation
|
| 266 |
+
|
| 267 |
+
Collected by the **UP Diliman Digital Signal Processing Laboratory**. This is a
|
| 268 |
+
repackaging for research use; the underlying corpus terms are those of UP-DSP.
|
| 269 |
+
**Please credit the original collectors**, and contact UP-DSP for terms covering
|
| 270 |
+
uses beyond research.
|
| 271 |
+
|
| 272 |
+
If you represent UP-DSP and want attribution, terms, or access changed, please
|
| 273 |
+
open a discussion on this repo.
|
| 274 |
+
|
| 275 |
+
---
|
| 276 |
|
| 277 |
+
## Contributing
|
|
|
|
| 278 |
|
| 279 |
+
Eight of these ten languages have essentially no public ASR or TTS baseline.
|
| 280 |
+
That is the opportunity here.
|
| 281 |
|
| 282 |
+
- Train a baseline on any single language and tag this dataset in your model card
|
| 283 |
+
- Report bad rows via the **Community** tab (include `source_file` and `session_id`)
|
| 284 |
+
- Improve the pipeline: [github.com/sapinsapin/halohalo](https://github.com/sapinsapin/halohalo)
|