NIST Speech Discs 10-1.1, 10-2.1
Developed by NYNEX Science and Technology
Produced on CD-ROM by the National Institute of Standards and Technology (NIST)
Copyright 1990 NYNEX Corporation
All Rights Reserved
The NTIMIT corpus was developed by the NYNEX Science and Technology Speech Communication Group to provide a telephone bandwidth adjunct to the popular TIMIT Acoustic-Phonetic Continuous Speech Corpus [NIST Speech Disc 1-1.1/NTIS PB91-505065, October 1990]. The corpus was contributed by NYNEX to NIST for distribution on CD-ROM media.
Please note that NYNEX retains full copyright on the corpus and all associated materials.
NTIMIT was collected by transmitting all 6300 original TIMIT utterances though various channels in the NYNEX telephone network and redigitizing them. Calibration tones were also recorded for each of the transmission channels.
The NTIMIT utterances were time-aligned with the original TIMIT utterances so that the TIMIT time-aligned transcriptions can be used with the NTIMIT corpus as well. See the documentation and publications listed below under "Related Publications" for more information on TIMIT and NTIMIT.
Unlike TIMIT, the NTIMIT corpus occupies 2 CD-ROMs. Please note that the data on the first NTIMIT disc, CD10-1.1, contains speech and auxiliary files which have a one-to-one correspondence with the original TIMIT CD-ROM files. The second NTIMIT disc, CD10-2.1, contains only the NTIMIT calibration tones. The online documentation has been duplicated on both discs for your convenience.
CD10-1.1 contains directories and files which correspond to the structure of the original TIMIT CD-ROM, the only difference is the "ntimit" vs. "timit" top-level directory name. The speech and associated data is organized on the CD-ROM according to the following hierarchy:
/ / / . where, USAGE :== train | test DIALECT :== dr1 | dr2 | dr3 | dr4 | dr5 | dr6 | dr7 | dr8 SEX :== m | f SPEAKER_ID :== where, TEXT_TYPE :== sa | si | sx SENTENCE_NUMBER :== 1 ... 2342 FILE_TYPE :== wav | txt | wrd | phn (see the file, "timit.doc", for information regarding the TIMIT corpus and its organization) (see the file, "ntimit.doc", and the accompanying documentation booklet for information regarding the NTIMIT corpus) Examples: /ntimit/train/dr1/fcjf0/sa1.wav (NTIMIT corpus, training set, dialect region 1, female speaker, speaker-ID "cjf0", sentence text "sa1", speech waveform file) /ntimit/test/df5/mbpm0/sx407.phn (NTIMIT corpus, test set, dialect region 5, male speaker, speaker-ID "bpm0", sentence text "sx407", phonetic transcription file)
CD10-2.1 contains the NTIMIT calibration tones. A set of tones exists for each transmission channel used. The calibration tones are organized on the CD-ROM according to the following hierarchy:
/ where, BIN :== bin01 | bin02 | ... | bin 10 LTU :== ltu0001 | ltu0002 | ... | ltu1274 (note: not all numbers are used) TONE :== long.wav | sweep.wav (see the file, "ntimit.doc", and the accompanying documentation booklet for an explanation of the calibration tones and their organization) Example: ntimit/calibrat/bin05/ltu0192/long.wav (NTIMIT corpus, calibration data, BIN 5, LTU 192, long tone)
The following documentation files are located in the "/ntimit/doc" directory on each of the discs:
NTIMIT: ntimit.doc - Overview of the NTIMIT corpus ltu_list.txt - A list of the LTUs used for the creation of utt_ndx.txt - Index of all 6300 TIMIT utterances and the BIN and LTU that they were transmitted to. TIMIT: timit.doc - Description of the TIMIT CD-ROM prompts.txt - Table of sentence prompts and sentence-ID numbers phoncode.doc - Table of phone symbols used in phonemic dictionary and phonetic transcriptions timitdic.doc - Description of the phonemic lexicon timitdic.txt - Phonemic dictionary of all orthographic words in prompts obj_phn.doc - Documentation for "obj_phn.txt" obj_phn.txt - Table of TIMIT objective phonetic characteristics spkrinfo.txt - Table of speaker attributes spkrsent.txt - Table of sentence-ID numbers for each speaker testset.doc - Description of suggested test/train subdivision