- a joint project between Texas Instruments(TI), MIT, and SRI.
- includes 6300 read sentences, with 10 sentences each from 630 speakers. The 6300 sentences were drawn from a set of 2342 sentences, some selected to have particular dialect shibboleths, others to maximize phonetic diphone coverage.
- each sentence in the corpus was phonetically hand-labeled, the sequence of phones was automatically aligned with the sentence wave file, and then the automatic phone boundaries were manually hand-correct.

TIMIT corpus

- consists of 3.5 hours of sentences extracted from the Switchboard corpus, together with transcriptions time-aligned at the syllable level. 


Switchboard Transcription Project

- a phonetically transcribed corpus of spontaneous American speech, containing about 300,000 words from 40 talkers.

The Buckeye corpus

- A collection of waveforms is hand-labeled with the corresponding string of phones. 

San Jose State University

Saint Olaf College

Phonetic Resources

An on-going but a helpful book resource about NLP
https://web.stanford.edu/~jurafsky/slp3/

Speech and Language Processing (3rd ed. draft) 

- It offers phonetic transcriptions for words.
- The Linguistic Dictionary Consortium (LDC) provides pronunciation lexicons for Egyptian Arabic, Dutch, English, German, Japanese, Korean, Mandarin, and Spanish.
- The open-source CMU Pronouncing Dictionary (CMU, 1993) has pronunciations for about 134,000 word forms, while the fine-grained 110,000-word UNISYN dictionary, freely available for research purposes, gives syllabifications, stress, and also pronunciations for dozens of dialects of English.

Online pronunciation dictionaries

Phonetically annotated corpus

The categorization of texts into organized groups, which is usually implemented by supervised learning.
Applications: sentiment analysis, spam detection, news categorization, topic analysis, and more.

Text Classification

The pre-training and fine-tuning paradigm is a method motivated by the goal of creating adaptable, general-purpose systems for universal language understanding and generation. It involves separating the common components of neural network architectures, such as Transformers, and training them on vast amounts of unlabeled data using self-supervision. The resulting systems, known as foundation models, can be easily adapted for specific downstream applications via fine-tuning or prompting. This paradigm shift has enormously transformed natural language processing, meaning that in many cases, large-scale supervised learning for specific tasks is no longer required.

Learn Before

Related

Learn After