Tiv-TTS-Dataset

Language

Tiv (ISO 639-3: tiv; also referred to as Zwa Tiv by its speakers) is a Benue-Congo language of the wider Niger-Congo family. It is classified within the Tivoid subgroup of the Southern Bantoid languages and is the largest language in that subgroup by number of speakers. Tiv is spoken primarily in the Middle Belt region of Nigeria, with significant speaker populations in neighbouring areas of Cameroon.

Tiv is spoken by an estimated 4 to 5 million first-language speakers, making it one of the major indigenous languages of central Nigeria. The majority of speakers are concentrated in Benue State, which is considered the historical and cultural heartland of the Tiv people, as well as in Taraba, Nasarawa, Plateau, and parts of Cross River and Adamawa States. Diaspora communities of Tiv speakers are also found across urban centres in Nigeria, including Abuja, Kaduna, Lagos, Jos, and Makurdi.

The language holds an important place in the cultural and political life of central Nigeria. It is used in education, local broadcasting, religious services, literature, and everyday communication, and it plays a central role in the identity of the Tiv people.

Variants

Tiv exhibits regional and lineage-based variation, though all varieties are broadly mutually intelligible. Dialectal differentiation in Tiv is traditionally organized along the lines of the major Tiv ancestral groupings (ityô), with each grouping associated with a particular geographic area and set of lexical, phonological, and prosodic tendencies.

Commonly recognized Tiv dialect groupings include:

• Ihyarev — spoken predominantly in the central-western Tiv-speaking area, including parts of Gwer, Gwer-West, and surrounding Local Government Areas of Benue State

• Shitire — spoken in the north-eastern Tiv region, around Katsina-Ala and adjacent areas

• Kparev — spoken widely in the central Tiv-speaking area, including areas of Gboko and its surroundings; often associated with the written standard

• Iharev — spoken in the southern and south-western Tiv-speaking area

• Ukum — spoken in the north-eastern Tiv region

• Masev, Tombo, Nôngov, Ikurav, Iwuese, and other smaller groupings — spoken in additional Tiv traditional territories, each with distinctive lexical and phonological features

The variety represented in this dataset reflects the speech of a single speaker who identifies with the Ihyarev dialect. The written transcriptions follow the standard modern Tiv orthography as used in contemporary Tiv publications, education, and media, and are broadly intelligible across dialectal varieties.

Writing System

Tiv is written using a Latin-based orthography whose modern form was developed progressively from the early twentieth century onward. The earliest written records of the language were produced by Christian missionaries (notably Dr. Carl Cum of the Sudan United Mission, who began work on the Benue river in 1904, and Rev. A. S. Judd and Rev. W. A. Malherbe, who contributed to its codification). The orthography was subsequently revised and standardized through the work of the Tiv Language Committee (Mzoo u Zwa Tiv) and, in the later twentieth century, in collaboration with the Nigerian Educational Research and Development Council (NERDC) and the College of Education, Katsina-Ala.

Key orthographic features of written Tiv:

• Tiv uses the Latin alphabet extended with the character ô to represent a distinct vowel quality; accented vowels such as á, é, í, ó, ú and the character ú are also used to mark tone or vowel distinctions in certain publications.

• Tiv is a tonal language with two main tones (high and low) and a downstep; tone is marked in pedagogical and linguistic publications, but is often omitted in general-purpose written material.

• Words are typically written disjunctively, with prefixes, stems, and modifiers separated by spaces, in contrast to the conjunctive orthographies of many southern Bantu languages.

• Vowel sequences (diphthongs), geminate consonants, and labialized/palatalized consonants are represented by digraphs (e.g., "gh", "kp", "gb", "hw", "ngw", "ny", "sh", "ts").

The transcriptions in this dataset follow the standard modern Tiv orthography as used in contemporary Tiv written publications. Tone is generally not marked except where the source text includes it.

Grammar and Linguistic Features

Tiv is an agglutinative Benue-Congo (Tivoid) language with a rich verbal and nominal morphology. Key grammatical features include:

Noun class system:

• Tiv organizes nouns into a reduced noun-class system inherited from Niger-Congo, typically analysed as having five to seven classes distinguished by prefixes (e.g., "i-", "u-", "a-", "mba-", "ma-"), each with associated singular/plural pairings.

• Concordial agreement between nouns and their dependents (adjectives, possessives, demonstratives, numerals, verbal subject markers) is expressed through class-appropriate agreement prefixes, though the system is less elaborated than in most southern Bantu languages.

Verbal morphology:

• The Tiv verb hosts subject markers, tense/aspect/mood markers, negation, and various extensions (including applicative, causative, reciprocal, and stative forms). Auxiliaries and serial verb constructions are also widely used.

• Tense and aspect distinctions include present, past, future, perfective, imperfective, habitual, and anterior, expressed through combinations of particles, auxiliaries, and verbal morphology.

Phonology:

• Tiv has a moderately rich consonant inventory that includes labial-velar stops (kp, gb), prenasalized consonants, and labialized/palatalized consonants.

• It is a tonal language with high and low tones and a downstep, which participate in both lexical and grammatical distinctions.

Word order:

• Basic word order is Subject–Verb–Object (SVO), with considerable flexibility supported by serial verb constructions, topicalization, and focus strategies.

Source

The textual material in this dataset is drawn from Tiv-language written sources. These are traditional Tiv oral-narrative material committed to writing including:

• Ethnographic and historical texts

• Proverbs

• Riddles

• Children's and folk stories

• Oral narratives

• Explanatory / didactic texts

The texts were segmented into short utterances suitable for read speech and used as prompts for audio recording sessions.

Domain

This dataset is derived from prompted read speech. The speaker read aloud pre-written Tiv texts drawn from traditional oral-narrative sources. The content covers a range of registers typical of contemporary written Tiv, including expository prose, folk narrative, proverbial and riddling genres.

The dataset has been structured as segmented, read-style speech suitable for speech synthesis tasks.

Size

The dataset is composed of 14 folders containing audio clips and corresponding mapping files.

Each folder contains between 172 and 175 audio files. Individual audio clips typically range from 1 to 30 seconds in duration.

Folder-level durations range from approximately 21 minutes 14 seconds to over 29 minutes of audio.

The dataset represents a total of 2,443 audio files with a combined duration of approximately 5 hours 50 minutes and 43 seconds of segmented Tiv speech.

A detailed breakdown of durations and file counts per folder is provided below.

Structure

Each folder in the dataset contains:

• A collection of audio files in MP3 format

• A tab-separated mapping file linking each audio file to its transcription

Each line in the mapping file follows the format:

audio_filename.mp3 key sentence attempts

The dataset is designed for TTS pipelines requiring paired audio-text data.

Sample

Below are representative entries drawn from one of the mapping files in the dataset, illustrating the pairing between audio filenames and their Tiv transcriptions:

1. db9633fe11996fa8c10adf8eef468ea9.mp3 | Nahan lu un voul er shi ve mba imôngo yô.

2. 9e33f6c5b8835b1a45dfff3c15f2f73c.mp3 | Kasev mba hen iya u Ageva mba ungwan er Adan- Wade va yô

3. cdfc8801ca56bf68e0dceca7947a7d70.mp3 | Tuv u hiihii hen igba na la, Adan-Wade fatyb yaven fefa ga.

4. fba2f3f34e39b3a3561d8820dc3975e9.mp3 | nahan do u nzuvul a mbayev ga" Verbima u mbayev ve zua a kwayan tsembelee ve vanthan wanye alu sha

5. d7218322ef57173edc95384ebd0f88ab.mp3 | Er shi ngula va hingir u maren wan nahan kpa ior shi kaa er lu tom na lor ôron akaa

6. af483e852cf0a13685e35757141ab9f8.mp3 | Mbajima Se me nenge Wan môm môm Se me nenge Tsô Adan-Wade hingir wanye-kwaor yo

7. bc6dbf40ee1216ad305b632a9c6dced1.mp3 | Ve lu vinen amar ior lu namben ve inyarev shi yôron ve iwer sôngon ilev.

8. be2c7d907ac7a1d5199a6686e263ae52.mp3 | Do ve hen geri shon kpishi sha aciu ior ngohol ve her do.

9. 31e13831c38c1876966762a568ef89b1.mp3 | Hiihii la nasera teman tswen u eren akaa ne.

10. 3acc56bc422a37f16be8cc3a88c0687e.mp3 | ver hen kpeifi nahan agirgir a va kula.