neh7777/Pretraining-V1

Indic TTS Unified v1

A large-scale, unified collection of speech data for text-to-speech (TTS) and speech research. This dataset consolidates 17 distinct source datasets into a single, schema-normalized resource covering Indian / South Asian languages, plus major European, African, MENA, and Central Asian languages, with over 13.7 million utterances and 26,000+ hours of audio.

All audio is resampled to 24 kHz mono. Every row follows an identical schema regardless of source, enabling seamless multi-dataset training without per-source preprocessing.

---

Dataset Summary

Statistic	Value
Total utterances	13,777,541
Total audio duration	26,300+ hours
Languages covered	58+
Audio format	24 kHz, mono, float32
Configs (subsets)	26

---

Subsets

Config	Rows	Hours	Speakers	Languages	Source Dataset
orpheus_distill_neucodec	400	~2	--	1	BarryFutureman/orpheus-distill-neucodec
maya_distill_neucodec	14,000	33.6	12,801	1	BarryFutureman/maya-distill-data-neucodec
emodb_neucodec	22,043	40.3	5	1	BarryFutureman/EmoDB-neucodec
expresso_neucodec	11,599	10.9	4	1	BarryFutureman/expresso-neucodec
nonverbal_tts	6,222	17.6	2,296	1	deepvk/NonverbalTTS
elise	1,194	2.6	1	1	MrDragonFox/Elise
elise_hindi	1,147	2.4	1	1	ronith09/Elise-Hindi
synthetic_v1	10,759	22.8	50	9	kenpath/tts-synthetic-v1
spicor	50,468	99.4	2	1	kenpath/tts-SPICOR
indictts	294,008	527.0	151,247	14	SPRINGLab/IndicTTS (14 datasets)
msft_indian	115,392	134.9	115,390	3	deepdml/microsoft-speech-corpus-indian
syspin	786,625	1,706.5	18	9	kenpath/tts-SYSPIN
ivr	664,208	1,656.5	10,152	22	ai4bharat/indicvoices_r
rasa	582,195	1,035.5	40	22	ai4bharat/Rasa
kathbath	805,721	1,475.2	985	12	ai4bharat/Kathbath
shrutilipi	2,226,753	4,665.0	--	16	ai4bharat/Shrutilipi
cv22_sidon	3,212,858	4,614.2	--	17	sarulab-speech/commonvoice22_sidon
cv22_african	725,125	~1,500	--	6	sarulab-speech/commonvoice22_sidon (African subset: sw, lg, ha, yo, ig, am)
cv22_central_asian	404,288	~800	--	4	sarulab-speech/commonvoice22_sidon (Central Asian subset: uz, ka, az, kk)
cv22_mena	194,077	~370	--	2	sarulab-speech/commonvoice22_sidon (MENA subset: ar, fa)
cv22_de	699,462	~1,450	--	1	sarulab-speech/commonvoice22_sidon (German)
cv22_fr	700,202	~1,450	--	1	sarulab-speech/commonvoice22_sidon (French)
cv22_es	1,592,537	~3,300	--	1	sarulab-speech/commonvoice22_sidon (Spanish)
cv22_european	591,663	~1,200	--	6	sarulab-speech/commonvoice22_sidon (European subset: it, nl, tr, ru, pt, pl)
arabic_misc	49,412	122.7	39,898	1	Mixed: MohamedRashad, Nourhann, NeoBoy, saleh1312, KejueAI
uq_speech	16,183	28.0	16,183	1	ixxan/mms-tts-uig-script_arabic-UQSpeech
libritts_r	358,000	585.0	2,456	1	parler-tts/libritts_r_filtered
Total	14,135,541	26,885+		58+

---

Schema

All configs share the same column schema:

Column	Type	Description
audio	Audio (24 kHz)	Audio waveform, resampled to 24 kHz mono
text	string	Transcript text. Rasa transcripts may include emotion tags (see below)
speaker_id	string	Speaker identifier (see Speaker ID Policy below)
source	string	Name of the originating dataset (e.g., "kathbath", "rasa")
language	string	Full language name (e.g., "Hindi", "Bengali", "Tamil")
gender	string	"Male", "Female", or "Unknown"
duration	float64	Audio duration in seconds

Speaker ID Policy

Speaker identification varies by source dataset:

• Deterministic 8-character hash: For datasets that provide speaker metadata (kathbath, syspin, ivr, rasa, spicor, synthetic_v1, cv22_sidon, cv22_african, cv22_central_asian, cv22_mena, cv22_de, cv22_es, cv22_fr, cv22_european), the speaker_id is a deterministic hash derived from the original speaker label, ensuring consistency across rows from the same speaker.
• Random UUID: For datasets without reliable speaker metadata (shrutilipi, msft_indian, indictts), each row receives a unique random UUID. These should not be used for speaker-level grouping.

Duration

Duration values are unfiltered -- no minimum or maximum duration threshold (such as 0.5--60s) has been applied. Downstream consumers should apply their own filtering as needed.

Emotion Tags (Rasa)

The rasa config contains expressive/emotional speech. Transcript text in this subset may include inline emotion tags such as <happy>, <sad>, <angry>, <surprise>, <fear>, <disgust>, and <neutral>. These tags indicate the intended emotion of the utterance and can be used for emotion-conditioned TTS training.

---

Language Coverage

The dataset spans a broad range of Indian languages. The table below lists languages and the configs in which they appear:

Language	Configs
Assamese	shrutilipi, ivr, rasa, cv22_sidon
Bengali	kathbath, shrutilipi, ivr, rasa, syspin, cv22_sidon
Bodo	ivr, rasa
Dhivehi	cv22_sidon
Dogri	shrutilipi, ivr, rasa
Dutch	cv22_european
English (Indian)	spicor, ivr, rasa, indictts
Arabic	arabic_misc, cv22_mena
French	cv22_fr
German	cv22_de
Italian	cv22_european
Polish	cv22_european
Portuguese	cv22_european
Russian	cv22_european
Spanish	cv22_es
Turkish	cv22_european
Uyghur	uq_speech
English (Common Voice)	cv22_sidon
Gujarati	kathbath, shrutilipi, ivr, rasa, syspin, indictts
Hindi	kathbath, shrutilipi, ivr, rasa, syspin, msft_indian, indictts, synthetic_v1, cv22_sidon
Kannada	kathbath, shrutilipi, ivr, rasa, syspin, indictts
Kashmiri	ivr
Konkani	shrutilipi, ivr, rasa
Maithili	shrutilipi, ivr, rasa
Malayalam	kathbath, shrutilipi, ivr, rasa, syspin, indictts, cv22_sidon
Manipuri	ivr, rasa
Marathi	kathbath, shrutilipi, ivr, rasa, syspin, indictts, cv22_sidon
Nepali	shrutilipi, ivr, rasa, cv22_sidon
Odia	kathbath, shrutilipi, ivr, rasa, syspin, indictts, cv22_sidon
Pashto	cv22_sidon
Punjabi	kathbath, shrutilipi, ivr, rasa, syspin, indictts, cv22_sidon
Rajasthani	indictts
Sanskrit	kathbath, shrutilipi, ivr, rasa
Santali	ivr, cv22_sidon
Saraiki	cv22_sidon
Sindhi	ivr, cv22_sidon
Tamil	kathbath, shrutilipi, ivr, rasa, syspin, msft_indian, indictts, cv22_sidon
Telugu	kathbath, shrutilipi, ivr, rasa, syspin, msft_indian, indictts, cv22_sidon
Urdu	kathbath, shrutilipi, ivr, rasa, cv22_sidon

---

Detailed Subset Descriptions

orpheus_distill_neucodec

Decoded from BarryFutureman/orpheus-distill-neucodec, an Orpheus distillation dataset stored as NeuCodec tokens. Contains 400 English utterances (~2 hours) with emotion conditioning. Text includes emotion wrapper tags (e.g., <happy>...</happy>) and converted vocal expression tags (e.g., <sigh>, <laugh>). Speaker IDs are random 8-character hex values (no speaker metadata in source).

maya_distill_neucodec

Decoded from BarryFutureman/maya-distill-data-neucodec, a Maya distillation dataset stored as NeuCodec tokens. Contains 14,000 English utterances (33.6 hours) with emotion conditioning and rich voice metadata. Text includes emotion wrapper tags (e.g., <happy>...</happy>) and vocal expression tags (e.g., <giggle>, <sigh>, <yawn>). Speaker IDs are deterministic 8-character hashes derived from voice_description, yielding 12,801 unique speakers. Gender breakdown: Male 4,602, Female 4,681, Unknown 4,717.

emodb_neucodec

Decoded from BarryFutureman/EmoDB-neucodec, a synthetic emotional speech dataset with GPT-4o-generated English text and NeuCodec-encoded audio. Contains 22,043 utterances (40.3 hours) after deduplication, with 5 speakers and 7 emotion styles (angry, happy, sad, fearful, surprised, disgusted, neutral). Text includes emotion wrapper tags (e.g., <angry>...</angry>). Speaker IDs are deterministic 8-character hashes of the speaker name. All gender values are "Unknown".

expresso_neucodec

Decoded from BarryFutureman/expresso-neucodec, the Expresso corpus encoded as NeuCodec tokens. Contains 11,599 English utterances (10.9 hours) after deduplication, with 4 speakers and multiple expressive styles. Text includes style wrapper tags (e.g., <confused>...</confused>). Speaker IDs are deterministic 8-character hashes of the original speaker ID (e.g., ex01). All gender values are "Unknown".

nonverbal_tts

Sourced from deepvk/NonverbalTTS, a nonverbal-annotated speech dataset combining Expresso and VoxCeleb data. Contains 6,222 English utterances (17.6 hours) with 2,296 unique speakers. Text uses the annotated Result column which includes emoji markers for nonverbal cues (e.g., 🌬️ for breath, 😤 for exhale). Emotion wrapping applied only for happy and sad categories. Gender breakdown: Male 3,872, Female 2,350.

elise

Sourced from MrDragonFox/Elise, a single-speaker English female dataset. Contains 1,194 utterances (2.6 hours). Audio resampled from 22050 Hz to 24 kHz. Text passed through as-is (includes emotion expression tags).

elise_hindi

Sourced from ronith09/Elise-Hindi, a Hindi version of the Elise dataset with the same speaker. Contains 1,147 utterances (2.4 hours). Audio resampled from 22050 Hz to 24 kHz.

synthetic_v1

Synthetic TTS data generated for bootstrapping and augmentation. Covers 9 languages (primarily Hindi) with 50 distinct synthetic voices. 10,759 utterances totaling 22.8 hours.

spicor

The SpiCor corpus of Indian English read speech. Contains 50,468 utterances (99.4 hours) from 2 speakers. Useful for high-quality single-speaker or few-speaker English TTS.

indictts

Derived from the SPRINGLab/IndicTTS collection, which spans 14 individual language datasets. Contains 294,008 utterances (527.0 hours) across 14 Indian languages. Speaker IDs are random UUIDs (no original speaker metadata available).

msft_indian

Sourced from deepdml/microsoft-speech-corpus-indian. Covers 3 languages (Hindi, Tamil, Telugu) with 115,392 utterances (134.9 hours). Speaker IDs are random UUIDs.

syspin

The SYSPIN TTS dataset provides high-quality studio-recorded speech across 9 languages from 18 speakers. With 786,625 utterances and 1,706.5 hours, this is one of the largest single-source contributions. Well-suited for single-speaker and multi-speaker TTS due to consistent recording conditions.

ivr

Derived from ai4bharat/indicvoices_r (IndicVoices-R), a large-scale read speech corpus. Covers 22 languages with 664,208 utterances (1,656.5 hours) from 10,152 speakers. One of the most linguistically diverse configs in this collection.

rasa

The ai4bharat/Rasa dataset of expressive and emotional Indian language speech. Covers 22 languages with 582,195 utterances (1,035.5 hours) from 40 speakers. Transcripts include inline emotion tags (e.g., <happy>, <sad>, <angry>) that indicate the expressed emotion, making this subset uniquely valuable for emotion-conditioned TTS.

kathbath

Sourced from ai4bharat/Kathbath, a read speech dataset covering 12 Indian languages. Contains 805,721 utterances (1,475.2 hours) from 985 speakers.

Language breakdown by hours:

Language	Hours
Tamil	176.9
Marathi	152.0
Kannada	150.3
Telugu	146.7
Hindi	139.6
Malayalam	139.1
Punjabi	128.4
Gujarati	113.4
Bengali	88.0
Odia	81.8
Sanskrit	80.4
Urdu	78.6

Gender breakdown: Female 982.7h, Male 492.5h

cv22_sidon

Sourced from sarulab-speech/commonvoice22_sidon, a SIDON-processed variant of Mozilla Common Voice 22.0. A curated selection of 17 South Asian / Indic language configs is included, covering all splits (train, validation, test, other, invalidated) merged into a single train split per config. Contains 3,212,858 utterances totaling 4,614.2 hours.

Speaker IDs are deterministic 8-character SHA256 hashes of the original Common Voice client_id (preserves speaker grouping across utterances while anonymizing).

Language breakdown:

Language	Code	Rows	Hours
English	en	1,687,562	2,670.8
Bengali	bn	957,937	1,129.8
Tamil	ta	181,715	314.2
Urdu	ur	201,883	244.8
Pashto	ps	57,167	79.3
Odia	or	23,329	36.5
Dhivehi	dv	23,875	33.3
Sindhi	sd	25,011	29.1
Hindi	hi	16,250	22.7
Marathi	mr	10,836	19.1
Malayalam	ml	9,121	10.7
Assamese	as	4,656	7.6
Saraiki	skr	5,825	6.7
Punjabi	pa-IN	3,136	4.2
Telugu	te	2,290	2.6
Nepali	ne-NP	1,416	1.6
Santali	sat	849	1.1

Processing pipeline: raw Common Voice audio (typically MP3 at 32--48 kHz) was decoded, downmixed to mono, and resampled to 24 kHz using high-quality resampling. All splits per language were concatenated. Gender values are mapped from the original gender field (male_masculine → Male, female_feminine → Female, otherwise Unknown).

cv22_de

German (de) Common Voice 22, sourced from sarulab-speech/commonvoice22_sidon. Contains 699,462 utterances (~1,450 hours). Same processing pipeline as cv22_sidon. Speaker IDs are deterministic 8-character SHA256 hashes of the original Common Voice client_id.

cv22_fr

French (fr) Common Voice 22, sourced from sarulab-speech/commonvoice22_sidon. Contains 700,202 utterances (~1,450 hours). Same processing pipeline as cv22_sidon. Speaker IDs are deterministic 8-character SHA256 hashes of the original Common Voice client_id.

cv22_es

Spanish (es) Common Voice 22, sourced from sarulab-speech/commonvoice22_sidon. Contains 1,592,537 utterances (~3,300 hours) across 320 train shards. Same processing pipeline as cv22_sidon. Speaker IDs are deterministic 8-character SHA256 hashes of the original Common Voice client_id.

cv22_european

A combined config of mid-size European Common Voice 22 languages, sourced from sarulab-speech/commonvoice22_sidon. Contains 591,663 utterances (~1,200 hours) across 6 languages: Italian (it), Dutch (nl), Turkish (tr), Russian (ru), Portuguese (pt), Polish (pl). Same processing pipeline as cv22_sidon. Speaker IDs are deterministic 8-character SHA256 hashes of the original Common Voice client_id.

shrutilipi

The largest config, sourced from ai4bharat/Shrutilipi. Contains 2,226,753 utterances (4,665.0 hours) across 16 languages: Assamese, Bengali, Dogri, Gujarati, Hindi, Kannada, Konkani, Maithili, Malayalam, Marathi, Nepali, Odia, Punjabi, Sanskrit, Tamil, and Telugu. No speaker metadata is available -- each row has a unique UUID as speaker_id, and all gender values are "Unknown".

---

Usage

Load a specific config

from datasets import load_dataset

ds = load_dataset("kenpath/indic-tts-unified-v1", "kathbath", split="train")
print(ds[0])
# {'audio': {'path': ..., 'array': array([...]), 'sampling_rate': 24000},
#  'text': '...', 'speaker_id': 'a1b2c3d4', 'source': 'kathbath',
#  'language': 'Tamil', 'gender': 'Female', 'duration': 5.32}

Streaming mode (recommended for large configs)

from datasets import load_dataset

ds = load_dataset(
    "kenpath/indic-tts-unified-v1", "shrutilipi",
    split="train", streaming=True
)

for example in ds:
    audio_array = example["audio"]["array"]
    text = example["text"]
    # Process as needed
    break

Filter by language

from datasets import load_dataset

ds = load_dataset(
    "kenpath/indic-tts-unified-v1", "ivr",
    split="train", streaming=True
)

hindi_ds = ds.filter(lambda x: x["language"] == "Hindi")

for example in hindi_ds:
    print(example["text"])
    break

Load multiple configs

from datasets import load_dataset, concatenate_datasets

configs = ["kathbath", "syspin", "rasa"]
datasets = []
for config in configs:
    ds = load_dataset(
        "kenpath/indic-tts-unified-v1", config, split="train"
    )
    datasets.append(ds)

combined = concatenate_datasets(datasets)
print(f"Combined: {len(combined)} rows")

Duration filtering

from datasets import load_dataset

ds = load_dataset(
    "kenpath/indic-tts-unified-v1", "syspin",
    split="train", streaming=True
)

# Keep only utterances between 1 and 30 seconds
filtered = ds.filter(lambda x: 1.0 <= x["duration"] <= 30.0)

---

Data Processing

The following normalization steps were applied uniformly across all source datasets during construction:

1. Audio resampling: All audio resampled to 24 kHz mono using high-quality resampling.
2. Schema alignment: Every source dataset was mapped to the unified 7-column schema described above.
3. Speaker hashing: Where speaker labels were available, they were converted to deterministic 8-character hashes for privacy and consistency. Where unavailable, random UUIDs were assigned.
4. Split merging: Train and test splits from source datasets were combined into a single train split per config.

---

Intended Use

This dataset is designed for:

• Text-to-speech (TTS) model training across Indian languages
• Automatic speech recognition (ASR) pretraining and fine-tuning
• Speaker verification and speaker embedding research (for configs with reliable speaker IDs)
• Multilingual and cross-lingual speech research
• Emotion-conditioned speech synthesis (using the rasa config)

---

Limitations

• Speaker IDs for shrutilipi, msft_indian, and indictts are random UUIDs and do not represent actual speaker groupings. Do not use these for speaker-level analysis.
• Gender metadata is "Unknown" for the entire shrutilipi config and may be incomplete in other configs.
• Duration is unfiltered. Some utterances may be very short (sub-second) or very long. Apply duration filtering for TTS training.
• Text quality varies across sources. Some transcripts may contain noise, transliteration inconsistencies, or incomplete sentences.
• Emotion tags in Rasa are embedded in the transcript text and need to be parsed or stripped depending on the downstream task.

---

Citation

If you use this dataset, please cite the original source datasets as appropriate:

• IndicTTS: SPRINGLab/IndicTTS
• Kathbath: ai4bharat/Kathbath
• SYSPIN: kenpath/tts-SYSPIN
• IndicVoices-R: ai4bharat/indicvoices_r
• Rasa: ai4bharat/Rasa
• Shrutilipi: ai4bharat/Shrutilipi
• Microsoft Speech Corpus Indian: deepdml/microsoft-speech-corpus-indian
• SpiCor: kenpath/tts-SPICOR
• Common Voice 22 (SIDON): sarulab-speech/commonvoice22_sidon (derived from Mozilla Common Voice Corpus 22.0, CC-0)

---

License

Please refer to the individual source dataset licenses. This unified collection is provided under CC-BY-4.0 for the aggregation and schema normalization work. The underlying audio and text data retain the licenses of their respective sources.