Pronunciation Dictionary of speech recognition

Source: Internet
Author: User
1. Concept

The pronunciation Dictionary (lexicon) contains a mapping between words (words) and phonemes (phones), which is used to connect acoustic models and language models.

The position of the pronunciation dictionary in the speech recognition process is as shown in the figure:

A pronunciation dictionary contains a collection of words that the system can handle and indicates its pronunciation. By using the pronunciation dictionary to obtain the mapping relation between the modeling unit of the acoustic model and the modeling unit of the language model, the acoustic model and the language model are connected to form a search state space for decoding work.
   2. Dictionary Generation

The following procedure applies only to pronunciation dictionaries for Chinese speech recognition. 2.1 Pinyin-the mapping of phonemes

The first thing you need to determine is the translation rule/mapping relationship between pinyin and phonemes.
can have different mapping relations, such as the Pinyin "yi1" can correspond to "II I1", can also correspond to "y I1" (the former is the use of Tsinghua Speech recognition rules).
Different mapping relationships can have different recognition effects. 2.2 Chinese words-Pinyin mapping

You need to list as many Chinese words and their corresponding pinyin as possible, with multiple phonetic characters to list their different combinations. 2.3 g2p Tool Implementation

Through the above two steps can be achieved in Chinese word-phoneme conversion, that is g2p (Grapheme-to-phoneme Conversion).
The script can be used to input Chinese words, output the function of the corresponding phonemes. 2.4 Collecting Chinese words

The pronunciation dictionary needs to cover as many words as possible-phonemes.
The previous article constructs the language Model (II): Model Training and evaluation methods, the text corpus after the word statistics of the words and their word frequency, remove the low-frequency words and too long words, to get the Chinese vocabulary. 2.5 Generating Dictionaries

Using the Chinese vocabulary as input, you can get the word-phoneme mapping through the g2p tool, which is the pronunciation dictionary. 3. Reference article what is g2p? HTTP://WWW.VOXFORGE.ORG/HOME/DOCS/FAQ/FAQ/WHAT-IS-G2P Speech recognition Technology Introduction http://blog.csdn.net/rfc2008/article/details/ 9151755?utm_source=tuicool&utm_medium=referral

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.