KnowBrainer Speech Recognition
Decrease font size
Increase font size
Topic Title: Olympus WS-853 mp3 file transcription error
Topic Summary:
Created On: 07/31/2020 01:39 PM
Status: Post and Reply
Linear : Threading : Single : Branch
 Olympus WS-853 mp3 file transcription error   - skypix7 - 07/31/2020 01:39 PM  
 Olympus WS-853 mp3 file transcription error   - skypix7 - 07/31/2020 02:12 PM  
 Olympus WS-853 mp3 file transcription error   - Lunis Orcutt - 08/01/2020 02:23 AM  
 Olympus WS-853 mp3 file transcription error   - skypix7 - 08/01/2020 02:06 PM  
 Olympus WS-853 mp3 file transcription error   - Lunis Orcutt - 08/01/2020 10:24 PM  
 Olympus WS-853 mp3 file transcription error   - kkkwj - 02/24/2021 12:10 AM  
 Olympus WS-853 mp3 file transcription error   - Lunis Orcutt - 02/24/2021 04:48 PM  
Keyword
 07/31/2020 01:39 PM
User is offline View Users Profile Print this message

Author Icon
skypix7
New Member

Posts: 3
Joined: 07/30/2020

Though the manual specifically says to use DNS mode for mp3 files to be transcribed by my Dragon Pro 15.3, the program keeps spitting out the same error:

"The format of file x.mp3 is invalid."

I've seen comments above about stereo mode etc. but the DNS mode is specifically set up for transcription.

Any ideas why this isn't working?

I've seen stellar reviews online for people who used the same recorder, driving fast in a convertible, with remarkable accuracy. I can't even get it to transcribe.

thanks for any insights. 

new member Jim

 



-------------------------
JIMBO
 07/31/2020 02:12 PM
User is offline View Users Profile Print this message

Author Icon
skypix7
New Member

Posts: 3
Joined: 07/30/2020

May answer my own question...found a post here from 2018 with a similar problem. I took his advice, and converted the mp3 file to a wav file. Also noticed in Audacity that the DNS mode in the Olympus recorder still records in stereo. This is undoubtedly what the problem was. I will attempt to record again in DNS mode but try to set to mono instead of stereo.
I'm new to the recorder so still getting through the manual.
NOTE: the conversion as the other poster mentioned, from mp3 to wav, worked just fine. I just deleted one of the tracks by splitting the stereo first in Audacity, erasing the track, then converting to wav. Transcription was perfect....well, perfect enough.

 



-------------------------
JIMBO

 08/01/2020 02:23 AM
User is online View Users Profile Print this message

Author Icon
Lunis Orcutt
Top-Tier Member

Posts: 39490
Joined: 10/01/2006

                    Welcome (See Mission Statement)

 

Unfortunately, the Internet is full of self-proclaimed geniuses who can barely run their equipment. If you don't compare digital recorder against other digital recorders and algorithms, you are simply not qualified to make a statement. These people were delighted because it worked, not because it worked as well as it should. If you have any doubts, feel free to contact any of these knowledgeable folks and ask them how to set up the AutoTranscribe Folder Agent. That will tell you just how qualified they are.

 

Dragon cannot transcribe a stereo recording and the term MP3 quite literally means compressed (not so good) audio. Compressed audio means that you're not getting everything. If you're digital recorder is going to be an important tool in your workflow, it should include DS2 algorithms which utilize a 12 to 1 lossless fidelity. You won't have that option on your recorder because it was designed more for generic work and recording music. WAV files are not ideal but they are better than MP3s. However, the best digital recorderutilize DS2 algorithms and some of them are fairly inexpensive. Because speech recognition is all we do, our website only carries recorders, like the Olympus DS-9500 which include DS2 algorithms. Professional transcription recorders don't even include a stereo option



-------------------------

Change "No" to "Know" w/KnowBrainer 2020
Trial Downloads
Dragon/Sales@KnowBrainer.com 
(615) 884-4558 ex 1



 08/01/2020 02:06 PM
User is offline View Users Profile Print this message

Author Icon
skypix7
New Member

Posts: 3
Joined: 07/30/2020

Thank you Lunis. Appreciate your take on this topic. DS2 recorder mode, that's a helpful essential to know. I can't justify $300 and up for a recorder now so I'll stumble along with the occasional writing notes I make en fielde, excuse my faux French and when my novel hits the big time, I'll buy every recorder you sell, one for each room of my McMansion.

Keep up the good work, glad to have found you after years of riding the sub-ground Nuance commuter train which makes those annoying $150 stops every year or two.

I do like 15.3. I use it with the same studio mic I use for audiobook narration, and get wonderful results after only 12 years of frequent frustration. 

 



-------------------------
JIMBO
 08/01/2020 10:24 PM
User is online View Users Profile Print this message

Author Icon
Lunis Orcutt
Top-Tier Member

Posts: 39490
Joined: 10/01/2006

Speaking of Ver. 15.3… 15.6 is coming out soon and it will be free. We do not have an ETA but Nuance never releases a patch or full incremental update unless the need is critical. 



-------------------------

Change "No" to "Know" w/KnowBrainer 2020
Trial Downloads
Dragon/Sales@KnowBrainer.com 
(615) 884-4558 ex 1

 02/24/2021 12:10 AM
User is offline View Users Profile Print this message

Author Icon
kkkwj
Top-Tier Member

Posts: 852
Joined: 11/05/2015

Actually, I have used my inexpensive little general-use Panasonic RR-XS410 recorders for many years to auto-transcribe files. It has two mic meshes on the top corners (left and right) and produces a stereo MP3 file. I can load the MP3 files into Audacity and clearly see differences in the waveforms of the two tracks (therefore, the tracks are not clones or derivatives of each other). So, I think Dragon can transcribe my stereo input MP3 files, without question. However, if Lunis is correct, Dragon must just be ignoring one of the two tracks. At any rate, my experience has been (for years, and with different Panasonic and Sony recorders) that Dragon can transcribe something useful from them all.

-------------------------

Win10/11/x64, AMD Ryzen 7 3700X/3950X, 64/128GB RAM, Dragon 15.3, SP 7 Standard, SpeechStart, Office 365, KB 2017, Dragon Capture, Samson Meteor USB Desk Mic, Amazon YUWAKAYI headset, Klim and JUKSTG earbuds with microphones, 3 BenQ 2560x1440 monitors, Microsoft Sculpt Keyboard and fat mouse

 02/24/2021 04:48 PM
User is online View Users Profile Print this message

Author Icon
Lunis Orcutt
Top-Tier Member

Posts: 39490
Joined: 10/01/2006

We are surprised that Dragon is even accepting your stereo recordings but it would be preferable to make a monophonic recording.

-------------------------

Change "No" to "Know" w/KnowBrainer 2020
Trial Downloads
Dragon/Sales@KnowBrainer.com 
(615) 884-4558 ex 1

Statistics
32213 users are registered to the KnowBrainer Speech Recognition forum.
There are currently 3 users logged in.
The most users ever online was 12124 on 09/09/2020 at 04:59 AM.
There are currently 399 guests browsing this forum, which makes a total of 402 users using this forum.

FuseTalk Standard Edition v4.0 - © 1999-2022 FuseTalk™ Inc. All rights reserved.