ALOFON corpus

Opis

The ALOFON corpus is one of the multimodal database of word recordings in English, available at http://www.modality-corpus.org/. The ALOFON corpus is oriented towards the recording of the speech equivalence variants. For this purpose, a total of 7 people who are or speak English with native speaker fluency and a variety of Standard Southern British English were recorded in various variants resulting from the speaker's country of origin.

Each speaker has uttered 300 words or short phrases specially selected by the cooperating phonologist to include appropriate phonetic variants. The created list, used for recordings, contains 300 items, which are a collection of English vowels and consonants, which are distributed in representative consonant contexts. The material has been divided into 10 sections of 30 phrases each, which were then spoken by the speakers. Therefore, there are 10 recording sessions for each speaker.

The ALOFON corpuse contains more than 140 audio files (48 kHz/16 bit sampling) with a total duration of more than 3.5 hours and more than 240 video recordings (Vicon system reference camera, 120 frames per second, 720p resolution and RGB camera, 50 frames per second, resolution 1080p), with a total duration of more than 7 hours and 140 files with Facial Motion Capture recordings made using the Vicon motion capture system (special reflective markers were carefully applied to faces, especially the mouth of the speakers).

The ALOFON database is currently the only body that emphasizes an allophonic approach to speech analysis representing three modalities, including unique recordings from the multi-camera motion capture system (FMC- Facial Motion Capture).

Recording setup

Sample stills (RGB camera) from the corpus

Plik z danymi badawczymi

ALOFON.ZIP

3.6 GB, S3 ETag 2e33f298a55bf1e7802fae718409a004-8, pobrań: 89

Hash pliku liczony jest ze wzoru
hexmd5(md5(part1)+md5(part2)+...)-{parts_count} gdzie pojedyncza część pliku jest wielkości 512 MB

Przykładowy skrypt do wyliczenia:
https://github.com/antespi/s3md5

pobierz

Informacje szczegółowe o pliku

Licencja:: Niestandardowa
przeczytaj

Informacje szczegółowe

Rok publikacji:

2020

Data zatwierdzenia:

2020-12-17

Data wytworzenia:

2018

Język danych badawczych:

angielski

Dyscypliny:

informatyka techniczna i telekomunikacja (Dziedzina nauk inżynieryjno-technicznych)
językoznawstwo (Dziedzina nauk humanistycznych)

DOI:

10.34808/7v2c-2y58

Finansowanie:

Metodyka i technologia polimodalnej alofonicznej transkrypcji mowy

Weryfikacja:

Politechnika Gdańska

Słowa kluczowe

Powiązane zasoby

publikacja Database of speech and facial expressions recorded with optimized face motion capture settings

Cytuj jako

Autorzy

wyświetlono 345 razy

Wyszukiwarka