English Speech Files

Nested
ralfherzog-20070822_vf19.zip
User: ralfherzog
Date: 8/22/2007 3:32 am
Views: 3774
Rating: 19
Hello!  The quality of this submission should be better than the quality of my previous submissions because this time I am using my external Andrea USB soundcard.  This USB soundcard is much better than my onboard sound card. 

Speaker Characteristics:

Gender: male;
Age range: adult;
Pronunciation dialect: general American.

Recording Information:

Microphone make: Sennheiser PC131;
Microphone type: noise canceling headset;
Audio card make: Andrea USB Adapter;
Audio card type: USB;
Audio Recording Software: Audacity 1.2.6;
O/S: Windows XP Professional.

File Info:

File type: FLAC;
Sampling rate: 48kHz;
Sample rate format: 16bit;
Number of channels: 1;
Audio Processing: no

vf19-01 It was beating and waiting in the ambush of those black pits
vf19-02 Something vastly more thrilling had come into it now
vf19-03 It took him half an hour to reach the edge of it
vf19-04 But there was no longer the mother yearning in his heart
vf19-05 Besides, had he not whipped the big owl in the forest
vf19-06 After all, it was simply a mistake in judgment
vf19-07 Had it struck squarely it would have killed him
vf19-08 The Indian even poked his stick into the thick ground spruce
vf19-09 Pebbles and dirt flew along with hair and fur
vf19-10 And he was filled with a strange and foreboding fear
vf19-11 It was steel, a fisher trap
vf19-12 OW, a wild dog, he growled
vf19-13 He was a pariah; a wanderer without a friend or a home
vf19-14 That is the strange part of it
vf19-15 His freshly caught furs he flung to the floor
vf19-16 For that reason Le Beau had chosen him to fight the big fight
vf19-17 In the crib the baby sat up and began to prattle
vf19-18 She obeyed, shrinking back with the baby in her arms
vf19-19 His teeth shut with a last click
vf19-20 It was over when he made his way through the ring of spectators
vf19-21 In a flash he was on his feet, facing him
vf19-22 He thought he saw a shudder pass through the Factor's shoulders
vf19-23 The moon had already begun its westward decline
vf19-24 They laughed like two happy children
vf19-25 He pulled, and the log crashed down to break his back
vf19-26 Fast, but endure
vf19-27 A little before dawn of the day following, the fire relief came
vf19-28 The Indian felt the worship of her warm in his heart
vf19-29 He drew in a deep breath as he looked at them
vf19-30 Then he shouted, Shut up
vf19-31 He changed his seat for a steamer reclining chair
vf19-32 On the far corner of the compound fence a hawk brooded
vf19-33 To these he gave castor oil
vf19-34 Hatred and murder and lust for revenge they possessed to overflowing
vf19-35 Sheldon glanced at the thermometer
vf19-36 I'll see to poor Hughie
vf19-37 Her gray eyes were flashing, and her lips were quivering
vf19-38 Also, I want information
vf19-39 Let them go out and eat with my boys
vf19-40 I, I beg pardon, he drawled

Copyright (C) 2007  Ralf Herzog

This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program; if not, see <http://www.gnu.org/licenses/>. 

--- (Edited on 8/22/2007 3:32 am [GMT-0500] by ralfherzog) ---

ralfherzog-20070822_vf19.zip ralfherzog-20070822_vf19.zip

Notice: many prompts in "English Speech Files" were adapted from the prompt files contained in the CMU_ARCTIC speech synthesis database, which were in turn derived from out-of-copyright texts from Project Gutenberg, by the FestVox project at the Language Technologies Institute at Carnegie Mellon University.

Re: ralfherzog-20070822_vf19.zip
User: kmaclean
Date: 8/22/2007 7:49 am
Views: 134
Rating: 34

Hi Ralf,

I spoke too soon ... Flac (version 1.1.2) on Linux works perfectly with this submission!!!Smile

I must have had a transmission problem when I downloaded your previous submission to Linux - never thought of downloading a second time on Linux (very stupid of me not to do so ...).  I downloaded the file again and it now works fine.

Therefore it seems that the crc check was basically telling me that there was a problem with the "download", and that I should try it again.

thanks for your help,

Ken 

--- (Edited on 8/22/2007 8:49 am [GMT-0400] by kmaclean) ---

--- (Edited on 8/22/2007 8:52 am [GMT-0400] by kmaclean) ---


Notice: many prompts in "English Speech Files" were adapted from the prompt files contained in the CMU_ARCTIC speech synthesis database, which were in turn derived from out-of-copyright texts from Project Gutenberg, by the FestVox project at the Language Technologies Institute at Carnegie Mellon University.

FTP/HTTP - No integrity check
User: ralfherzog
Date: 8/22/2007 8:44 am
Views: 154
Rating: 19
Hi Ken,

So your problem was that the zip file was corrupted during the download from the voxforge.org webserver to your Fedora Linux desktop computer.  Well, I think that is probably always the problem if you transmit a file via HTTP or FTP over the Internet that there is no warning if something went wrong during the transmission.  So you can get a corrupted file, and the software doesn't warn you.

For FTP check Wikipedia: http://tinyurl.com/3ccgya

"No integrity check on the receiver side. If a transfer is interrupted, the receiver has no way to know if the received file is complete or not."

Probably, there is the same problem with HTTP.  

Maybe it is good to use FLAC: each FLAC file has an integrated integrity checksum or hash:

"FLAC uses CRC checksums for identifying corrupted frames."

http://en.wikipedia.org/wiki/FLAC#Technical_details

So the use of FLAC guarantees that you will be warned if an audio file would have been corrupted during the transmission. 

--- (Edited on 8/22/2007 8:44 am [GMT-0500] by ralfherzog) ---


Notice: many prompts in "English Speech Files" were adapted from the prompt files contained in the CMU_ARCTIC speech synthesis database, which were in turn derived from out-of-copyright texts from Project Gutenberg, by the FestVox project at the Language Technologies Institute at Carnegie Mellon University.

Re: ralfherzog-20070822_vf19.zip
User: kmaclean
Date: 8/24/2007 7:40 pm
Views: 199
Rating: 19

Hi Ralph,

thanks for this submission!

Flac audio is now being processed automatically in the acoustic model creation scripts.  Sometime in the future, I hope to convert all the audio from wav to flac format.  Your last two submissions should be processed tonight.

One interesting thing, is that it seems that you get much better Flac compression ratios with audio recorded at higher sampling rates.

thanks,

Ken 

--- (Edited on 8/24/2007 8:40 pm [GMT-0400] by kmaclean) ---


Notice: many prompts in "English Speech Files" were adapted from the prompt files contained in the CMU_ARCTIC speech synthesis database, which were in turn derived from out-of-copyright texts from Project Gutenberg, by the FestVox project at the Language Technologies Institute at Carnegie Mellon University.

I get very good compression ratios
User: ralfherzog
Date: 8/25/2007 1:48 am
Views: 391
Rating: 21

Hi Ken,

At a quality of 48 kHz/16 bit, the speech files are compressed to a size of normally less than 40% compared to the original wav files.  I think that FLAC is a good choice. Cool At the moment, I am convinced that a recording quality of 48 kHz/16 bit should be sufficient.  

I found an interesting information regarding the question following question: "Which Sound Cards are compatible with NaturallySpeaking?"

Take a look at this hyperlink: http://www.itspeaking.com/speech-faqs.asp#4

They mention that the soundcards "Creative® Labs SoundBlaster® Live, Player and Platinum (card only)" are recommanded.  And what is even more interesting is the following information:

"To use Dragon NaturallySpeaking the card must support 11,025 Hz mono, 16-bit sound recording."

So, that means that Dragon NaturallySpeaking (a very good speech recognition software, which I am using) possibly needs a sound quality of only 11.025 Hz.  

If you are being asked, which soundcard you would recommend, those soundcards probably are a good and not too expensive choice (compared to professional recording cards).

--- (Edited on 8/25/2007 1:48 am [GMT-0500] by ralfherzog) ---

--- (Edited on 8/25/2007 4:52 am [GMT-0500] by ralfherzog) ---


Notice: many prompts in "English Speech Files" were adapted from the prompt files contained in the CMU_ARCTIC speech synthesis database, which were in turn derived from out-of-copyright texts from Project Gutenberg, by the FestVox project at the Language Technologies Institute at Carnegie Mellon University.

PreviousNext