This release of Hub-5 Mandarin training data consists of 42 calls
derived from the CallFriend Mandarin (language-id) collection. The
transcripts cover a contiguous 5-30 minute segment taken from a
recorded conversation lasting up to 30 minutes. All of these calls
are being designated as additional training data for the LVCSR project
in Mandarin.

This release contains speech files along with the necessary
documentation regarding audit information and demographic
information. These files are not compressed. You will find a
description of transcription conventions in the file "hub5_man.txt." 


Summary of contents:
___________________________

docs/
    README.txt		This file.
                         
    hub5_man.txt	Full description of the HUB-5 telephone
                        speech and transcript corpus for
                        Mandarin.

    call_ids.lst	A list of the IDs of each audio file
			in the corpus.

    callinfo.txt	Explanation of the audit information
			provided in "callinfo.tbl".

    callinfo.tbl	A list of audit information as 
			explained in "callinfo.txt", with
			information on number and sex of
			speakers and several sound quality
			judgements.

    headerinfo.txt	Explanation of the SPH header 
			information provided in "header_info.tbl".

    headerinfo.tbl	A table of the data that was originally
			in the SPH header for each audio file
			before they were converted.

    spkrinfo.txt	Explanation of the speaker demographic
			information provided in "spkrinfo.tbl".

    spkrinfo.tbl	A table of information provided about 
			the speakers involved in each phone
			call, such as age and hometown.

    wrd_dvsn.prn	Explanation of the segmentation
			practices used in the transcription
			of the audio files.

    alignment.txt	Explanation of the computational 
    			alignment of the transcriptions and 
			audio files at the monophone and word
			levels.

data/
    
    audio/		Directory containing recordings of
    			the phone conversations in WAV u-law
			encoding.

    transcripts/	

        mandarin/	The original transcriptions created 
			manually by native Mandarin speakers,
			in GB2312 encoding.

		pinyin/		Computationally translated Pinyin version
			of the transcriptions made using LDC 
			lexicons, in UTF-8 encoding.

    alignment/

		tab_separated/	Aligned version of the transcription with
			the audio, with columns for monophones and
			words in addition to time intervals.
	
		TextGrid/
	    
			phon-level/	Monophone-level alignment of conversations
				in TextGrid format.

			word-level/	Word-level alignment of conversations
				in TextGrid format.

METADATA:
------------

Total Duration:         18:45:29

Duration by language:
    - Mandarin          18:45:29

Calls per caller: 1

Unique Speakers: TOTAL 104
    - ma_4160: 2 speakers (1 F, 1 M)
    - ma_4198: 2 speakers (0 F, 2 M)
    - ma_4227: 2 speakers (2 F, 0 M)
    - ma_4233: 3 speakers (2 F, 1 M)
    - ma_4249: 2 speakers (2 F, 0 M)
    - ma_4257: 2 speakers (2 F, 0 M)
    - ma_4258: 2 speakers (2 F, 0 M)
    - ma_4270: 3 speakers (2 F, 1 M)
    - ma_4389: 2 speakers (1 F, 1 M)
    - ma_4426: 4 speakers (2 F, 2 M)
    - ma_4447: 2 speakers (2 F, 0 M)
    - ma_4559: 3 speakers (2 F, 1 M)
    - ma_4689: 2 speakers (2 F, 0 M)
    - ma_4972: 3 speakers (2 F, 1 M)
    - ma_5010: 3 speakers (2 F, 1 M)
    - ma_5195: 2 speakers (1 F, 1 M)
    - ma_5260: 2 speakers (0 F, 2 M)
    - ma_5347: 2 speakers (0 F, 2 M)
    - ma_5520: 2 speakers (0 F, 2 M)
    - ma_5530: 2 speakers (2 F, 0 M)
    - ma_5542: 2 speakers (1 F, 1 M)
    - ma_5560: 3 speakers (1 F, 2 M)
    - ma_5636: 2 speakers (0 F, 2 M)
    - ma_5653: 2 speakers (0 F, 2 M)
    - ma_5690: 3 speakers (2 F, 1 M)
    - ma_5715: 2 speakers (1 F, 1 M)
    - ma_5716: 5 speakers (2 F, 3 M)
    - ma_5731: 4 speakers (2 F, 2 M)
    - ma_5773: 4 speakers (2 F, 2 M)
    - ma_5779: 2 speakers (2 F, 0 M)
    - ma_5784: 2 speakers (0 F, 2 M)
    - ma_5800: 2 speakers (1 F, 1 M)
    - ma_5806: 4 speakers (2 F, 2 M)
    - ma_5821: 2 speakers (1 F, 1 M)
    - ma_5824: 2 speakers (0 F, 2 M)
    - ma_5904: 4 speakers (2 F, 2 M)
    - ma_5905: 2 speakers (0 F, 2 M)
    - ma_5906: 2 speakers (2 F, 0 M)
    - ma_5930: 2 speakers (2 F, 0 M)
    - ma_5949: 2 speakers (2 F, 0 M)
    - ma_5975: 2 speakers (1 F, 1 M)
    - ma_5982: 2 speakers (1 F, 1 M)