audio audioduration (s) 4 14.9 | source stringclasses 1
value | bak float32 3.64 4.25 | sig float32 2.65 3.81 | ovrl float32 2.34 3.62 | duration float32 4 14.9 |
|---|---|---|---|---|---|
aishell3 | 4.1708 | 3.7048 | 3.4281 | 4.966 | |
aishell3 | 4.2186 | 3.6866 | 3.4635 | 4.8 | |
aishell3 | 4.1605 | 3.5769 | 3.3185 | 9.889 | |
aishell3 | 4.1924 | 3.6255 | 3.3963 | 9.379 | |
aishell3 | 4.1752 | 3.5326 | 3.2787 | 5.906 | |
aishell3 | 4.1684 | 3.5319 | 3.2699 | 6.338 | |
aishell3 | 4.0805 | 3.58 | 3.3035 | 4.193 | |
aishell3 | 3.8422 | 3.6173 | 3.2187 | 5.001 | |
aishell3 | 3.8255 | 3.3044 | 2.9263 | 5.6 | |
aishell3 | 4.0979 | 3.3129 | 3.0459 | 6.244 | |
aishell3 | 3.8105 | 3.3357 | 2.9494 | 4.469 | |
aishell3 | 3.9302 | 3.1392 | 2.8198 | 4.31 | |
aishell3 | 4.0156 | 3.563 | 3.2527 | 5.461 | |
aishell3 | 3.9357 | 3.3252 | 2.9747 | 6.78 | |
aishell3 | 4.0221 | 3.6584 | 3.3312 | 4.5 | |
aishell3 | 4.0466 | 3.5776 | 3.2728 | 5.417 | |
aishell3 | 4.1846 | 3.7119 | 3.483 | 5.42 | |
aishell3 | 4.1191 | 3.6392 | 3.3673 | 5.58 | |
aishell3 | 3.9554 | 3.1527 | 2.8344 | 4.586 | |
aishell3 | 4.0926 | 3.4045 | 3.1271 | 5.277 | |
aishell3 | 4.1476 | 3.538 | 3.2975 | 12.685 | |
aishell3 | 4.0983 | 3.4933 | 3.2313 | 5.686 | |
aishell3 | 4.0921 | 3.3915 | 3.1205 | 14.859 | |
aishell3 | 4.1213 | 3.2151 | 2.9677 | 13.16 | |
aishell3 | 4.036 | 3.3735 | 3.0756 | 7.13 | |
aishell3 | 3.9693 | 3.6316 | 3.2848 | 4.144 | |
aishell3 | 3.9732 | 3.4284 | 3.1148 | 7.652 | |
aishell3 | 4.1234 | 3.5521 | 3.2955 | 5.14 | |
aishell3 | 3.9416 | 3.2464 | 2.9285 | 5.399 | |
aishell3 | 4.0531 | 3.403 | 3.1275 | 4.913 | |
aishell3 | 4.0724 | 3.4363 | 3.1725 | 4.744 | |
aishell3 | 4.1516 | 3.6305 | 3.3817 | 4.657 | |
aishell3 | 4.1198 | 3.5072 | 3.2551 | 4.914 | |
aishell3 | 4.1705 | 3.5502 | 3.3149 | 4.788 | |
aishell3 | 4.1523 | 3.6271 | 3.3859 | 5.133 | |
aishell3 | 4.0954 | 3.5679 | 3.2898 | 5.36 | |
aishell3 | 4.104 | 3.6006 | 3.3224 | 5.062 | |
aishell3 | 4.1353 | 3.6101 | 3.3515 | 5.118 | |
aishell3 | 4.0938 | 3.3713 | 3.0908 | 4.656 | |
aishell3 | 4.1344 | 3.6495 | 3.3993 | 5.941 | |
aishell3 | 4.0434 | 3.4514 | 3.1709 | 5.447 | |
aishell3 | 3.9922 | 3.47 | 3.1668 | 5.584 | |
aishell3 | 4.0706 | 3.4832 | 3.2053 | 5.777 | |
aishell3 | 3.913 | 3.2216 | 2.8968 | 4.833 | |
aishell3 | 4.0641 | 3.3325 | 3.0592 | 5.931 | |
aishell3 | 3.7091 | 3.6227 | 3.1493 | 4.83 | |
aishell3 | 4.064 | 3.7128 | 3.4093 | 4.83 | |
aishell3 | 4.0231 | 3.5601 | 3.2489 | 5.248 | |
aishell3 | 4.0452 | 3.4635 | 3.1816 | 6.555 | |
aishell3 | 4.1057 | 3.4111 | 3.1598 | 6.728 | |
aishell3 | 4.1389 | 3.5958 | 3.3314 | 11.91 | |
aishell3 | 3.8438 | 3.6179 | 3.2055 | 5.991 | |
aishell3 | 4.0882 | 3.6929 | 3.392 | 5.619 | |
aishell3 | 3.748 | 3.6279 | 3.1758 | 6.641 | |
aishell3 | 4.1056 | 3.4391 | 3.1737 | 4.115 | |
aishell3 | 4.0322 | 3.4278 | 3.1495 | 4.195 | |
aishell3 | 4.0863 | 3.4464 | 3.1785 | 4.382 | |
aishell3 | 3.9768 | 3.3684 | 3.0608 | 4.289 | |
aishell3 | 4.0902 | 3.2843 | 3.02 | 4.24 | |
aishell3 | 4.0433 | 3.5398 | 3.2509 | 4.462 | |
aishell3 | 3.7462 | 3.6669 | 3.2071 | 4.133 | |
aishell3 | 3.8651 | 3.6878 | 3.3018 | 4.226 | |
aishell3 | 3.8784 | 3.4727 | 3.0982 | 4.226 | |
aishell3 | 4.055 | 3.6119 | 3.309 | 4.551 | |
aishell3 | 4.0686 | 3.6219 | 3.3167 | 5.201 | |
aishell3 | 4.1465 | 3.524 | 3.2819 | 4.61 | |
aishell3 | 4.1287 | 3.5786 | 3.326 | 5.41 | |
aishell3 | 3.8508 | 3.5948 | 3.1735 | 5.387 | |
aishell3 | 4.0521 | 3.4524 | 3.1585 | 10.681 | |
aishell3 | 4.169 | 3.7045 | 3.4772 | 4.592 | |
aishell3 | 4.0072 | 3.4698 | 3.1756 | 4.55 | |
aishell3 | 4.1195 | 3.5524 | 3.3014 | 4.876 | |
aishell3 | 3.837 | 3.5063 | 3.1158 | 4.876 | |
aishell3 | 4.1518 | 3.6587 | 3.4203 | 6.29 | |
aishell3 | 4.0702 | 3.5969 | 3.3178 | 6.44 | |
aishell3 | 4.155 | 3.5611 | 3.3259 | 6.59 | |
aishell3 | 4.1362 | 3.6195 | 3.3687 | 4.07 | |
aishell3 | 4.0851 | 3.6103 | 3.3505 | 4.22 | |
aishell3 | 4.1103 | 3.5069 | 3.2567 | 4.13 | |
aishell3 | 4.0292 | 3.5745 | 3.2628 | 4.01 | |
aishell3 | 3.9155 | 3.5317 | 3.1554 | 4.226 | |
aishell3 | 3.8648 | 3.4286 | 3.0793 | 4.087 | |
aishell3 | 3.9919 | 3.5389 | 3.2112 | 4.505 | |
aishell3 | 4.1778 | 3.6956 | 3.4507 | 4.923 | |
aishell3 | 4.1624 | 3.7437 | 3.4922 | 4.69 | |
aishell3 | 4.0887 | 3.6265 | 3.3528 | 4.83 | |
aishell3 | 4.0584 | 3.4698 | 3.1924 | 4.876 | |
aishell3 | 4.1219 | 3.679 | 3.4109 | 4.69 | |
aishell3 | 4.1309 | 3.6223 | 3.3649 | 4.969 | |
aishell3 | 4.1605 | 3.6953 | 3.4579 | 5.155 | |
aishell3 | 4.0287 | 3.4312 | 3.1284 | 4.969 | |
aishell3 | 3.9255 | 3.3259 | 2.9856 | 4.551 | |
aishell3 | 3.7486 | 3.2989 | 2.8944 | 4.923 | |
aishell3 | 3.6722 | 3.1826 | 2.7576 | 5.155 | |
aishell3 | 4.0753 | 3.4385 | 3.1609 | 5.155 | |
aishell3 | 3.8361 | 3.1429 | 2.7838 | 4.551 | |
aishell3 | 4.0956 | 3.4654 | 3.1874 | 5.433 | |
aishell3 | 3.6973 | 3.3019 | 2.8526 | 4.551 | |
aishell3 | 3.8104 | 3.1814 | 2.8185 | 5.619 | |
aishell3 | 3.9156 | 3.3628 | 3.0154 | 5.666 |
TTS-Clean44k
A multilingual pool of verified-clean, wideband speech for training and evaluating speech restoration / text-to-speech (TTS) models. Every utterance is independently checked on two axes and stored as parquet with its per-utterance quality scores attached:
- Native sample rate ≥ 44.1 kHz — measured per file with
ffprobe, never trusting the source's advertised rate. Anything below 44.1 kHz is dropped. - DNSMOS P.835
bak≥ 3.644 — the background-noise MOS from the DNSMOS P.835 model. Only genuinely clean recordings pass.
The dataset was assembled as the clean teacher pool for Sidon call-centre speech restoration (the decoder is trained to reproduce its teacher, so the teacher must be genuinely clean and full-band), but it is broadly useful as a filtered multilingual TTS corpus.
Composition
28 language/source configs · 119,950 utterances · 208.7 h · utterance-weighted DNSMOS bak 4.036 / sig 3.484 / ovrl 3.203.
| # | config | utts | hours | bak | sig | ovrl |
|---|---|---|---|---|---|---|
| 1 | hifitts | 19,447 | 30.0 | 4.026 | 3.533 | 3.247 |
| 2 | uk_ireland_en | 14,925 | 28.0 | 4.077 | 3.572 | 3.313 |
| 3 | cv_zhcn | 8,948 | 15.4 | 3.952 | 3.466 | 3.145 |
| 4 | basque | 6,907 | 13.5 | 4.043 | 3.521 | 3.230 |
| 5 | galician | 5,343 | 10.0 | 4.055 | 3.521 | 3.248 |
| 6 | catalan | 4,018 | 9.0 | 3.977 | 3.433 | 3.127 |
| 7 | peruvian_es | 4,991 | 8.8 | 4.052 | 3.423 | 3.154 |
| 8 | south_african | 5,240 | 8.3 | 3.977 | 3.460 | 3.149 |
| 9 | kannada | 3,542 | 7.6 | 4.045 | 3.436 | 3.165 |
| 10 | gujarati | 3,789 | 7.4 | 4.079 | 3.489 | 3.226 |
| 11 | argentinian_es | 4,362 | 6.8 | 4.077 | 3.520 | 3.246 |
| 12 | colombian_es | 4,105 | 6.8 | 4.052 | 3.404 | 3.144 |
| 13 | chilean_es | 3,755 | 6.6 | 4.056 | 3.424 | 3.160 |
| 14 | tamil | 3,633 | 6.4 | 4.027 | 3.300 | 3.035 |
| 15 | nigerian_en | 2,687 | 5.1 | 4.102 | 3.548 | 3.299 |
| 16 | javanese | 2,906 | 4.2 | 4.004 | 3.467 | 3.163 |
| 17 | malayalam | 2,470 | 4.1 | 4.034 | 3.367 | 3.095 |
| 18 | aishell3 | 2,814 | 4.1 | 4.047 | 3.513 | 3.218 |
| 19 | telugu | 2,517 | 4.0 | 4.053 | 3.358 | 3.098 |
| 20 | venezuelan_es | 2,466 | 4.0 | 4.057 | 3.473 | 3.208 |
| 21 | burmese | 2,071 | 3.6 | 4.020 | 3.495 | 3.203 |
| 22 | sundanese | 2,116 | 3.5 | 3.999 | 3.443 | 3.140 |
| 23 | khmer | 2,098 | 3.2 | 4.092 | 3.448 | 3.202 |
| 24 | marathi | 1,400 | 2.8 | 4.088 | 3.433 | 3.186 |
| 25 | yoruba | 1,464 | 2.2 | 4.035 | 3.476 | 3.178 |
| 26 | nepali | 1,157 | 2.0 | 4.000 | 3.496 | 3.183 |
| 27 | puertorico_es | 516 | 0.9 | 4.050 | 3.443 | 3.169 |
| 28 | cv_ta | 263 | 0.4 | 3.855 | 3.116 | 2.780 |
Schema
Each config is a single train split with columns:
| column | type | description |
|---|---|---|
audio |
Audio(sampling_rate=48000) |
mono waveform, decoded to 48 kHz |
source |
string |
source/config name |
bak |
float32 |
DNSMOS P.835 background-noise MOS (≥ 3.644 for every row) |
sig |
float32 |
DNSMOS P.835 signal MOS |
ovrl |
float32 |
DNSMOS P.835 overall MOS |
duration |
float32 |
clip length in seconds |
Long recordings are chunked to ≤ 15 s; short utterances (≥ 4 s) are kept whole.
Usage
from datasets import load_dataset
# load one config (source)
ds = load_dataset("Scicom-intl/TTS-Clean44k", "uk_ireland_en", split="train")
print(ds[0]["audio"], ds[0]["bak"], ds[0]["duration"])
# or stream a large config
ds = load_dataset("Scicom-intl/TTS-Clean44k", "hifitts", split="train", streaming=True)
Sources
- OpenSLR high-quality TTS — Javanese (41), Sundanese (44), Tamil (65), Telugu (66), Malayalam (63), Marathi (64), Khmer (42), Nepali (43), Gujarati (78), Kannada (79), Burmese (80).
- OpenSLR crowdsourced — Argentinian/Chilean/Colombian/Peruvian/Puerto-Rican/Venezuelan Spanish (61/71/72/73/74/75), Catalan (69), Basque (76), Galician (77), Yoruba (86), Nigerian English (70), South-African English (32), UK & Ireland English (83).
- AISHELL-3 (Mandarin), Hi-Fi TTS (English, capped at 30 h for balance), Common Voice
(Mandarin
zh-CN, Tamilta).
What the gates rejected
The verification is strict on purpose — sources that only claim to be high-fidelity were dropped:
- Common Voice
id(Indonesian) andyue(Cantonese) — excluded entirely; every clip was below 44.1 kHz (Cantonese was uniformly 32 kHz). - VCTK — omitted here only because its available mirror was download-throttled, not for quality.
Nothing below the sample-rate or DNSMOS bar is included.
Licensing
Audio is redistributed from the upstream corpora listed above; each retains its original license (OpenSLR corpora are variously CC BY / CC BY-SA / CC0, AISHELL-3 is research-use, Hi-Fi TTS is CC BY 4.0, Common Voice is CC0). Consult the corresponding source before commercial use. The DNSMOS scores and the 48 kHz re-encoding are provided as-is.
- Downloads last month
- 1,116