timit_asr / README.md
julien-c's picture
julien-c HF staff
Align/fix license metadata info (#4613)
9ddd964
metadata
pretty_name: TIMIT
annotations_creators:
  - expert-generated
language_creators:
  - expert-generated
language:
  - en
license:
  - other
license_details: LDC-User-Agreement-for-Non-Members
multilinguality:
  - monolingual
size_categories:
  - 1K<n<10K
source_datasets:
  - original
task_categories:
  - automatic-speech-recognition
task_ids: []
paperswithcode_id: timit
train-eval-index:
  - config: clean
    task: automatic-speech-recognition
    task_id: speech_recognition
    splits:
      train_split: train
      eval_split: test
    col_mapping:
      file: path
      text: text
    metrics:
      - type: wer
        name: WER
      - type: cer
        name: CER

Dataset Card for timit_asr

Table of Contents

Dataset Description

Dataset Summary

The TIMIT corpus of read speech is designed to provide speech data for acoustic-phonetic studies and for the development and evaluation of automatic speech recognition systems. TIMIT contains broadband recordings of 630 speakers of eight major dialects of American English, each reading ten phonetically rich sentences. The TIMIT corpus includes time-aligned orthographic, phonetic and word transcriptions as well as a 16-bit, 16kHz speech waveform file for each utterance. Corpus design was a joint effort among the Massachusetts Institute of Technology (MIT), SRI International (SRI) and Texas Instruments, Inc. (TI). The speech was recorded at TI, transcribed at MIT and verified and prepared for CD-ROM production by the National Institute of Standards and Technology (NIST).

The dataset needs to be downloaded manually from https://catalog.ldc.upenn.edu/LDC93S1:

To use TIMIT you have to download it manually.
Please create an account and download the dataset from https://catalog.ldc.upenn.edu/LDC93S1
Then extract all files in one folder and load the dataset with:
`datasets.load_dataset('timit_asr', data_dir='path/to/folder/folder_name')`

Supported Tasks and Leaderboards

  • automatic-speech-recognition, speaker-identification: The dataset can be used to train a model for Automatic Speech Recognition (ASR). The model is presented with an audio file and asked to transcribe the audio file to written text. The most common evaluation metric is the word error rate (WER). The task has an active leaderboard which can be found at https://paperswithcode.com/sota/speech-recognition-on-timit and ranks models based on their WER.

Languages

The audio is in English. The TIMIT corpus transcriptions have been hand verified. Test and training subsets, balanced for phonetic and dialectal coverage, are specified. Tabular computer-searchable information is included as well as written documentation.

Dataset Structure

Data Instances

A typical data point comprises the path to the audio file, usually called file and its transcription, called text. Some additional information about the speaker and the passage which contains the transcription is provided.

{
    'file': '/data/TRAIN/DR4/MMDM0/SI681.WAV',
    'audio': {'path': '/data/TRAIN/DR4/MMDM0/SI681.WAV',
      		  'array': array([-0.00048828, -0.00018311, -0.00137329, ...,  0.00079346, 0.00091553,  0.00085449], dtype=float32),
      		  'sampling_rate': 16000},
    'text': 'Would such an act of refusal be useful?',
    'phonetic_detail': [{'start': '0', 'stop': '1960', 'utterance': 'h#'},
                        {'start': '1960', 'stop': '2466', 'utterance': 'w'},
                        {'start': '2466', 'stop': '3480', 'utterance': 'ix'},
                        {'start': '3480', 'stop': '4000', 'utterance': 'dcl'},
                        {'start': '4000', 'stop': '5960', 'utterance': 's'},
                        {'start': '5960', 'stop': '7480', 'utterance': 'ah'},
                        {'start': '7480', 'stop': '7880', 'utterance': 'tcl'},
                        {'start': '7880', 'stop': '9400', 'utterance': 'ch'},
                        {'start': '9400', 'stop': '9960', 'utterance': 'ix'},
                        {'start': '9960', 'stop': '10680', 'utterance': 'n'},
                        {'start': '10680', 'stop': '13480', 'utterance': 'ae'},
                        {'start': '13480', 'stop': '15680', 'utterance': 'kcl'},
                        {'start': '15680', 'stop': '15880', 'utterance': 't'},
                        {'start': '15880', 'stop': '16920', 'utterance': 'ix'},
                        {'start': '16920', 'stop': '18297', 'utterance': 'v'},
                        {'start': '18297', 'stop': '18882', 'utterance': 'r'},
                        {'start': '18882', 'stop': '19480', 'utterance': 'ix'},
                        {'start': '19480', 'stop': '21723', 'utterance': 'f'},
                        {'start': '21723', 'stop': '22516', 'utterance': 'y'},
                        {'start': '22516', 'stop': '24040', 'utterance': 'ux'},
                        {'start': '24040', 'stop': '25190', 'utterance': 'zh'},
                        {'start': '25190', 'stop': '27080', 'utterance': 'el'},
                        {'start': '27080', 'stop': '28160', 'utterance': 'bcl'},
                        {'start': '28160', 'stop': '28560', 'utterance': 'b'},
                        {'start': '28560', 'stop': '30120', 'utterance': 'iy'},
                        {'start': '30120', 'stop': '31832', 'utterance': 'y'},
                        {'start': '31832', 'stop': '33240', 'utterance': 'ux'},
                        {'start': '33240', 'stop': '34640', 'utterance': 's'},
                        {'start': '34640', 'stop': '35968', 'utterance': 'f'},
                        {'start': '35968', 'stop': '37720', 'utterance': 'el'},
                        {'start': '37720', 'stop': '39920', 'utterance': 'h#'}],
    'word_detail': [{'start': '1960', 'stop': '4000', 'utterance': 'would'},
                    {'start': '4000', 'stop': '9400', 'utterance': 'such'},
                    {'start': '9400', 'stop': '10680', 'utterance': 'an'},
                    {'start': '10680', 'stop': '15880', 'utterance': 'act'},
                    {'start': '15880', 'stop': '18297', 'utterance': 'of'},
                    {'start': '18297', 'stop': '27080', 'utterance': 'refusal'},
                    {'start': '27080', 'stop': '30120', 'utterance': 'be'},
                    {'start': '30120', 'stop': '37720', 'utterance': 'useful'}],

    'dialect_region': 'DR4',
    'sentence_type': 'SI',
    'speaker_id': 'MMDM0',
    'id': 'SI681'
}

Data Fields

  • file: A path to the downloaded audio file in .wav format.

  • audio: A dictionary containing the path to the downloaded audio file, the decoded audio array, and the sampling rate. Note that when accessing the audio column: dataset[0]["audio"] the audio file is automatically decoded and resampled to dataset.features["audio"].sampling_rate. Decoding and resampling of a large number of audio files might take a significant amount of time. Thus it is important to first query the sample index before the "audio" column, i.e. dataset[0]["audio"] should always be preferred over dataset["audio"][0].

  • text: The transcription of the audio file.

  • phonetic_detail: The phonemes that make up the sentence. The PHONCODE.DOC contains a table of all the phonemic and phonetic symbols used in TIMIT lexicon.

  • word_detail: Word level split of the transcript.

  • dialect_region: The dialect code of the recording.

  • sentence_type: The type of the sentence - 'SA':'Dialect', 'SX':'Compact' or 'SI':'Diverse'.

  • speaker_id: Unique id of the speaker. The same speaker id can be found for multiple data samples.

  • id: ID of the data sample. Contains the .

Data Splits

The speech material has been subdivided into portions for training and testing. The default train-test split will be made available on data download.

The test data alone has a core portion containing 24 speakers, 2 male and 1 female from each dialect region. More information about the test set can be found here

Dataset Creation

Curation Rationale

[Needs More Information]

Source Data

Initial Data Collection and Normalization

[Needs More Information]

Who are the source language producers?

[Needs More Information]

Annotations

Annotation process

[Needs More Information]

Who are the annotators?

[Needs More Information]

Personal and Sensitive Information

The dataset consists of people who have donated their voice online. You agree to not attempt to determine the identity of speakers in this dataset.

Considerations for Using the Data

Social Impact of Dataset

[More Information Needed]

Discussion of Biases

[More Information Needed]

Other Known Limitations

Dataset provided for research purposes only. Please check dataset license for additional information.

Additional Information

Dataset Curators

The dataset was created by John S. Garofolo, Lori F. Lamel, William M. Fisher, Jonathan G. Fiscus, David S. Pallett, Nancy L. Dahlgren, Victor Zue

Licensing Information

LDC User Agreement for Non-Members

Citation Information

@inproceedings{
  title={TIMIT Acoustic-Phonetic Continuous Speech Corpus},
  author={Garofolo, John S., et al},
  ldc_catalog_no={LDC93S1},
  DOI={https://doi.org/10.35111/17gk-bn40},
  journal={Linguistic Data Consortium, Philadelphia},
  year={1983}
}

Contributions

Thanks to @vrindaprabhu for adding this dataset.