 |
KnowBrainer Speech Recognition |
 |
 |  |
|
Topic Title: Transcribe from command line Topic Summary: Created On: 02/26/2022 03:07 AM Status: Post and Reply |
|
 02/26/2022 03:07 AM
|
|
|
Is it possible to transcribe recordings from the command line? (I know about "Tools/Transcribe Recording", but would like to batch this) |
|
|
|
|
|
02/26/2022 07:44 AM
|
|
|
"Tools/Transcribe Audio/AutoTranscribe Folder Agent" already does that for you.
-------------------------
The New Game in Town: DragonConnect |
|
|
|
|
|
02/26/2022 12:53 PM
|
|
|
Just adding a wee bit to Rüdiger's sage advice… After you set up the AutoTranscribe Folder Agent, you can drop it into your Startup folder and have it open minimized (out of the way but always available). The AutoTranscribe Folder Agent cannot transcribe recordings when Dragon is open but if you close Dragon, the the AutoTranscribe Folder Agent will automatically transcribe your recordings; without prompts. Setting up this process is detailed in our KnowBrainer Dragon Installation/Training Guide which additionally pictorially details fixing 28 sandtraps in your user profile and is complementary to our Dragon customers. ------------------------- Change "No" to "Know" w/KnowBrainer 2022 |
|
|
|
|
|
02/26/2022 04:32 PM
|
|
|
Another recommendation (based on LONG experience transcribing many hundreds of mp3 files into Word documents) is to use the "transcribe to text" option instead of transcribing into Word. First, the Word transcriptions take longer to load Word each time. Second, if you're like me and want to join the many little mp3 transcriptions into one or more larger documents, that's FAR harder to do in Word, even you program it like I did. In contrast, transcribing into TXT files is far faster and then you can use any convenient editor (or other tools) to join and/or process the TXT files when you join them.
------------------------- 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, excellent Sareville Wireless Mono Headset, 3 BenQ 2560x1440 monitors, Microsoft Sculpt Keyboard and Logitech G502 awesome gaming mouse. |
|
|
|
|
|
02/26/2022 09:00 PM
|
|
|
You are wasting your time. In order to join multiple files into one, don't use the AutoTranscribe folder agent, but go with "Transcribe recording" instead, where can select multiple files in one go so they will be appended to each other automatically. When finished, choose to edit the transcription in DragonPad, which will default to RTF, and keep the DRA until you're done, as you may want to go back to the audio for when ever recognition doesn't come out as expected. -------------------------
The New Game in Town: DragonConnect |
|
|
|
|
|
02/27/2022 05:01 PM
|
|
|
If you are using the AutoTranscribe Folder Agent to transcribe multiple recordings directly from your recorder, it should transcribe each of your recordings, one at a time, into the same document. In other words, a single document with all of your transcriptions. We discovered this by accident ------------------------- Change "No" to "Know" w/KnowBrainer 2022 |
|
|
|
|
|
02/27/2022 06:43 PM
|
|
|
Addendum: It appears that when using the AutoTranscribe Folder Agent that you can no longer play back embedded dra recordings from within your document (random-access playback). We haven't used this in a while but wonder if it is broken in Ver. 15. Fortunately, we are able to randomly playback text if we use Dragon's Transcribe Recording menu option ------------------------- Change "No" to "Know" w/KnowBrainer 2022 |
|
|
|
|
|
02/27/2022 10:12 PM
|
|
|
It must depend on the format, or possibly the filename conventions of the recordings.
As you can see from the excerpt of some of my transcriptions below, I get a separate .txt file for each .m4a.
2020-11-08_10-48-09.m4a
The iPhone app Just Press Record creates filenames of the form 10-48-09.m4a or 10-48-09.wav, in directories by day-date. I flatten those into the date and time stamp filenames, while moving them in batches from iCloud to a PC directory for the auto transcription agent. typically a batch of several days' to weeks' recordings at a time.
In any case, I like them as separate files, since they usually end up as separate OneNote pages. ------------------------- DPG15.6 (also DPI 15.3) + KB, Sennheiser MB Pro 1 UC ML, BTD 800 dongle, Windows 10 Pro, MS Surface Book 3, Intel Core i7-1065G7 CPU @ 1.3/1.5GHz (4 cores, 8 logical, GPU=NVIDIA Quadro RTX 3000 with Max-Q Design. |
|
|
|
|
|
03/04/2022 11:29 PM
|
|
|
RW: That's an interesting idea - select all the mp3 files and choose Transcribe Recording. Thank you! I will try that next time. I usually do little recordings over a period of a month and then transcribe them all at once.
Lunis: I didn't know that Dragon could transcribe directly from a recorder. I have always dragged them from the recorder to a Windows folder where they get transcribed. ------------------------- 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, excellent Sareville Wireless Mono Headset, 3 BenQ 2560x1440 monitors, Microsoft Sculpt Keyboard and Logitech G502 awesome gaming mouse. |
|
|
|
|
|
03/05/2022 03:57 AM
|
|
|
But make sure to do it the other way round. Go to "Transcribe Recording", and in the "Browse" window, select multiple files at once. -------------------------
The New Game in Town: DragonConnect |
|
|
|
|
FuseTalk Standard Edition v4.0 - © 1999-2023 FuseTalk™ Inc. All rights reserved.