Click here to return to the main FAQ page.
Click here for an introduction to
transcribing music, and also read the Overview and Getting Started sections of the Help that comes with
Transcribing music is hard work and although Transcribe! makes it easier, it is still hard work.
If you have never transcribed music from recordings at all then it can sometimes be a good idea to start by using traditional methods - using a perfectly ordinary music player for your first transcriptions. Then try Transcribe! again. That way, you learn one thing at a time instead of trying to learn everything at once.
Also see the query below, "Does Transcribe! actually do the transcribing?".
Full documentation is in the Help that comes with Transcribe! itself.
Start with the Overview and Getting Started sections.
The FAQ page you are reading now has a lot of information.
And you will find various videos showing how to use Transcribe! near the bottom of this page.
No. There are (and have been since the 1970's) some very clever people working on this with
very limited success on anything but the simplest music. You can find links to
other people's efforts here.
My view is that it's too hard to do well, and useless if done badly.
Transcribe! does attempt to guess what notes are being played at any chosen moment (marking them on the piano keyboard) and also to name the chords being played. But it makes no attempt to deal with rhythm, or to produce musical notation or midi.
Also see here for a further reflection of mine on this subject.
No, Transcribe! does not address the question of notating the transcription - some
people use notation software, some don't write anything down at all (they are
learning to play something by playing along). Personally I use pencil and paper first,
possibly entering it into a music notation program later.
You can add textual annotations (see the Text menu) to the transcription, useful for lyrics, chord symbols, comments etc.
I don't plan to add notational capability (standard notation, tablature, Music XML, or whatever) to Transcribe! - if I give it limited capability then people will immediately start complaining about the things it can't do, while full-featured music notation is a huge job, and there are already any number of software packages around for this, click here.
Yes, Transcribe! can display video synchronised to the audio at any speed. If the video is online (e.g. YouTube etc.) then you will need to download it first, and you may need to convert the video to a suitable format as well. See Help - Various Topics - Video, and also see this page.
No, Transcribe! is for transcribing music from sound recordings. There are any number of software packages around for MIDI, click here.
If you simply want to keep the part you are interested in and remove the rest then you can do this
by selecting (with the mouse) the part you want and using the Export Sound File (or Export Video)
command to create a new file containing just that part.
But for anything more complicated the answer is No, Transcribe! is for transcribing music from recordings. There are any number of software packages around for recording, editing and mixing sound, click here.
Yes, Transcribe! can be used very effectively for transcribing speech. There is some advice about this in Transcribe!'s help, under the heading "Various Topics".
Yes, Transcribe! can be used very effectively for play-along practice. There is some advice about this in Transcribe!'s help, under the heading "Various Topics".
Zoom/Skype etc. have clever noise-cancelling algorithms which attempt to remove background noise,
and also to prevent the sound from the user's loudspeaker feeding back into the microphone.
This can mean that if you play sound from Transcribe!, the person at the other end may not hear it.
Solutions to try are:
(1) On Zoom there is an option to "share computer sound". On Skype if you select Screen Sharing then there should be an option to also share sound. Search online for "Zoom/Skype share computer sound" for more information.
(2) The person playing sound in Transcribe! should use headphones for hearing Zoom/Skype sound but tell Transcribe! to output through the speakers (Preferences - Playback) so it gets picked up by the mic.
(3) In Zoom/Skype turn off whatever automatic adjustment and others options there may be for the mic. E.g. in Zoom - Settings - Audio there is an option to "Turn On Original Sound". In Skype you can turn off Settings - "Automatically adjust microphone settings".
(4) Turn up the volume of Transcribe! playback.
After you pause playback with the spacebar, if you press the spacebar again you will play again from the current point (the little red triangle). To position the current point, click on the waveform - there is more about this in the FAQs just below. If you want to continue playback from the paused point instead, then you need to un-pause. You can do this with the comma or semicolon keys (see Help - File menu - Keyboard Shortcuts), or the pause button in the lower toolbar.
If you want the spacebar to resume paused playback instead of starting again from the current point then you could use Keyboard Shortcuts on the File menu to have the spacebar trigger the command "PlayResume". If you do this then it might make sense to program some other keystroke to trigger "PlaySpacebar" as it can be useful to have a keystroke which starts playback from the current point.
Transcribe! 9.20 has the <enter> key programmed by default to do this. If you have updated from a previous version of Transcribe! then you would need to press the "Default" button in the Keyboard Shortcuts dialog, to load the new default shortcuts.
To set the current point (the red triangle which marks the place where playback will begin) : click on the waveform.
(If there is already a selection there then you will need to shift-click or double-click instead, or you will find you
are adjusting the existing selection instead).
To make a selection (e.g. for looping) : sweep the mouse over the waveform with the button pressed, so the desired section is selected (highlit).
To adjust one end of an existing selection : click and drag, near the end you want to adjust. You can also use the arrow keys, for details see the Help for Keyboard Shortcuts (File menu).
To move an existing selection : click within the existing selection. You can also use the square bracket keys for moving the current point or selection, for details see the Help for Keyboard Shortcuts.
If you hold the shift key down then this will change the behaviour - you can tell what is going to happen by the shape of the cursor before you click. Also double-click will always collapse an existing selection. As also will the 'U' keyboard shortcut.
In fact the little red triangles aren't really markers in the usual sense, they just indicate the current point, which is where playback will begin if you press the Play button or the spacebar. And since playback always has to begin somewhere, you can't delete the current point. See the query just above about positioning the current point and making selections, and see Help - Getting Started for more info. Also there is a "Back to the beginning" button in the lower toolbar which will move the current point to the start of the track.
Transcribe! does not have commands specifically for moving the paused playback point, though of course any form
of playback will do it (regular playback, Cue & Review, and "scrubbing").
When people ask this question it is because they are trying to use the paused playback point as their "current point"
but that's not really how Transcribe! was designed.
You should instead use the current point (the red triangle) as your current point and then it will all make more sense. There are lots of commands for moving the the current point and selection, some configured as keyboard shortcuts by default and many more you can create shortcuts for if you want. One particular shortcut that is relevant here is 'E' which sets the current point to the currently playing or paused point.
Another option that you might try is Preferences - Playback - "Paused playback point follows selection" (first appeared in Transcribe! 9.0). Press the Help button on that Preferences page for more info.
If you want to preserve a specific point or selection in order to recall it later then you should use stored loops for this (on the Misc page of the Fx window). There are also many commands for using stored loops & selections, which you can create shortcuts for if you want.
Also see the query below, "How can I start playback at a specific point without changing the current point or selection?".
First, be aware that "markers" have nothing to do with looping. You choose
what to loop by selecting it with the mouse. So:
- On the "Play" menu, make sure that "Loop" is selected - in fact this is the default anyway, as long as you haven't changed it.
- Select the section you want to loop - that is, sweep the mouse over the waveform display with the button pressed, so the desired section is selected (highlit).
- Press the Play button on the toolbar. That's it.
You can also do it while playing by pressing the "A-B Loop" button on the toolbar : once to set the loop start point and then again to select the loop end.
You can store and recall any number of loops, on the Misc page of the Fx window. It can also be useful to create keyboard shortcuts for recalling loops.
This is where you listen to the track at one point (i.e. a very short loop), and then
move that point around. The way to do it is to set a short loop playing - see the query above about looping -
and then click and drag (or use the square-bracket shortcut keys) to move that short loop around. See the Help for Keyboard
Shortcuts (File menu) for details about these commands.
If you turn on the Spectrum (View menu), then you will see the Spectrum of the selection you have made (and optionally Note and Chord guesses), which will change as you move the selection around.
There is also a "scrub" playback mode. See Help - Menus and Toolbar - Mouse Commands, for details.
You can control-click (command-click on Mac) on the waveform at the required start point.
Also if you show the Navigation Bar & the Time Line (View menu) then you can click on the Time Line to start from any point in the track.
If you have selected a section for looping and then use this technique to start playback before the loop then when playback reaches the loop, it will loop that section.
Remember also that if you want to play other parts of a track without losing a loop you have carefully set up, then the answer is to save the loop (on the Misc page of the Fx window). Then you can restore that loop any time you want.
Press the "reset to start" button in the lower toolbar (below and to the right of the Stop button), then the Play button also in the lower toolbar.
If you want to make it even easier, you can create keyboard shortcuts - the command is on the File menu. Try this:
On key press: SelectionResetStart (in the Selection category)
On key release: PlaySelectionStart (in the Playing category)
Alternatively, there is also a command called "PlayAll".
There is a command which will do this, it's called PlayLoopRelease. You can program a keyboard shortcut if you want to use it, it's in the Loop category.
My preferred way of marking a section of music is to play it through once, tapping the 'M' key at the start of
each measure. If you want the beats marked then double-click the first marker and specify auto
subdivisions - I very seldom use actual beat markers. And if you want the sections
marked then double-click (or right-click) the relevant measure marker and convert it to a section
marker (or use 'S' instead of 'M' for marking it in the first place).
Remember that the Navigation Bar (View menu) allows you to go directly to any marker.
If you want to mark events that happen in arbitrary places (e.g. the place where a particular lyric or chord or note happens) then Text Blocks are usually more suitable - see the Help for more about Text Blocks.
Every now and then, someone proposes a scheme for automatically generating markers at fixed
intervals, or for converting Transcribe!'s automatic subdivision markers into "real" markers.
However I have so far resisted such suggestions as they would add to the complexity of using
markers, and I am not at all convinced that they would offer significant benefit. My preferred
way of marking a section of music is described in the query just above. This is very simple and it
gets the measures marked in the same time it takes to play them. If you have some more complex
scheme in mind then please ask yourself, is it really any quicker or easier than the method I
have just outlined?
Having said that, there is a command on the Markers menu called "Equally Space Selected Markers" which can be used for placing a number of equally spaced markers between two existing markers. For example, suppose you have placed two section markers and want to divide that section into eight measures by adding seven measure markers. With playback stopped (not paused), click anywhere between the section markers to position the current point (the little red triangle), hold down <shift> and tap 'M' seven times to place seven measure markers on top of each other at the current point. Then use the mouse to select a region from the first section marker to the second one, including both of them, and use the "Equally Space Selected Markers" command to spread out the measure markers you placed.
In Help - Menus & Toolbar - File Menu - Keyboard Shortcuts, you can read about keyboard shortcuts
such as shift-[ and shift-] which will take you to the previous or next beat marker.
In Help - Getting Started - The Transcription Window, you can read about the Navigation Bar which you can right-click (control-click on Mac) to get a pop up menu where you can jump directly to a chosen marker.
And on the Misc page of the Audio Effects and Controls window you will find buttons to take you to the next or previous marker.
You can, but it's a two step process. First you must create a stored loop marking the place in question (a stored loop can be simply a position). Click on the relevant point of the waveform, then store that point as a loop on the Misc page of the Fx window (click the Help button on that page for more details).
Then you can create a shortcut for one of the commands LoopRecall1, LoopRecall2, etc as appropiate, in the Looping category.
Let's suppose that you have placed some beat or measure markers in your transcription.
You can create or adjust a selection with the mouse (see above) or with the keyboard
(see the Help for Keyboard Shortcuts) but how about if you want to adjust the selection to start or end
exactly on a beat or measure marker? The simplest way is to shift-click the marker triangle, which will extend the selection to that marker.
Another way is described in Help - Getting Started - The Transcription Window :
"If you hold down the <alt> key when clicking or dragging then the selection will 'snap' to the nearest marker if
there is one near. The <alt> key may not work if it is used to bring up a system menu instead. In that case you
can instead press <control> after you have started making your selection, for the same effect."
You can also set up keyboard shortcuts to do this kind of thing if you want, though they are not configured by default because there are so many possibilities to choose from. Transcribe! has many commands which are not immediately visible. This is deliberate as we don't want to scare new users with a complex interface.
Read the Help for Keyboard Shortcuts which tells you how to configure any command to be triggered by a keystroke of your choice. You will see a group of commands with names like "SelectionIncreaseRightMeasure" which increases the current selection at the right by one measure, and so on.
For example: to select a whole measure around the current point, configure a keyboard command with its "press" command as "SelectionIncreaseRightMeasure" and its "release" command as "SelectionIncreaseLeftMeasure". Then to advance the selection to the next whole measure, program another key with its "press" command as "SelectionIncreaseRightMeasure" and its "release" command to "SelectionDecreaseLeftMeasure".
Another example: to "select all", configure a keyboard command with its "press" command as "SelectionMoveLeftEnd" and its "release" command as "SelectionIncreaseRightEnd".
These commands can also be triggered from foot pedals of course.
Also see the "Move" buttons on the "Misc" page of the Audio Effects and Controls window. Read the Help for that page, to find out more.
There is a special command for this, see Preferences - Playback - "Playback Start Delay", and press the Help button on that page for details about how to use it.
Of course if you have a pedal then you can start playback with that, possibly eliminating the need for a delay.
There are default kb shortcuts for jumping by measures or sections but not by seconds, however you can easily create them.
If you want to move the current point (the red triangle) by one second then program a shortcut as:
On Press: SelectionMoveRightMs1000 (or left of course)
but if you want to jump from the currently playing or paused point then:
On Press: SelectionMakeZero (to move the current point to the currently playing point)
On Release: SelectionMoveRightMs1000 (to move the current point)
Press the Help button in the Keyboard Shortcuts dialog for more info.
Transcribe! does not have a specific playlist feature, however it's not difficult to get the same effect. When you Save a transcription this creates a transcription (xsc) file which is a small text file containing information about the transcription, including the name and location of the sound file. The xsc file does not need to be in the same folder as the sound file it uses, and you can move xsc files to different folders however you like (as long as they are not open in Transcribe! at the time). So, put the xsc files of your "playlist" tracks into a single folder (named "My practice playlist" or whatever) and then you can open them all at once by selecting them all in your file browser and drag-n-drop them onto the Transcribe! window.
If a tune appears in more than one list then it's fine to copy the transcription file - transcription files are small. For more discussion of this technique, please see Help - Various Topics - Organising Many Transcriptions.
It may be that you have placed section and measure markers throughout a track and now you want to copy them to another transcription - another sound file - which has exactly the same structure. This is quite common with stem files where the stems are in separate sound files. Suppose your current transcription file is "stem1.xsc" and the sound file is "stem1.mp3". Now you want to create a "stem2.xsc" for sound file "stem2.mp3", and having the same markers. First use Transcribe's "Save" command to save "stem1.xsc" if there are unsaved changes. Then use the "Save As" command to save the transcription again, as "stem2.xsc". Now use the "Import Sound File" command to import "stem2.mp3", thus replacing the sound file associated with "stem2.xsc". That's it.
Update July 2023 the latest version of Transcribe! (9.3) has special handling for stem files, which makes the above tricks unnecessary. If you are running 9.3 then see the Stem Files command on the File menu, and press the Help button in that dialog.
You can download some example stem files here.
If you want to save a modified sound or video file from Transcribe! (with altered speed or pitch for instance) then
the command you need is called "Export Sound File" (or "Export video") (on the File menu).
XSC (transcription) files don't contain any sound. They are text files containing information about your transcription. Read the Help for the File Menu (in Transcribe!'s Help) for more information.
The idea about exporting modified sound (or video) files from Transcribe! is that first you set up the modifications you
want using the Fx (Audio Effects and Controls) window, so that the file is playing in the same way that you want to export it.
Then use the Export command and check the boxes for the effects you want applied to the exported file.
So, if the speed change you applied is not appearing in the exported file then it probably means you didn't check the "Speed" checkbox in the Export dialog. On the other hand if the checkbox in the Export dialog is disabled, this is because you didn't apply the effect in the first place - cancel the Export command and use the Fx window to apply the effect you want, then Export again.
Assuming that you do, in fact, have enough disk space then the reason is that you are trying to save the file to an inappropriate location such as your system folder, or a CD drive which can't be recorded to anyway. Select an appropriate location such as somewhere in your Documents folder, and all will be well.
Transcribe! on Windows and Mac reads mp4, m4a & aac files but not m4p which is a protected format (usually from the
Apple music store). However, Apple stopped using this format in early 2009 and these days all iTunes music downloads
from the Apple music store are "iTunes Plus". "iTunes Plus" is a higher quality download and more to
the point, it comes without DRM (digital rights management) - it's an "m4a" file
rather than an "m4p". This means you can play it on non-Apple devices and software, including Transcribe!
You can upgrade previously downloaded tracks to "iTunes Plus". You can find Apple's page telling you how to do this by searching for "About iTunes Plus", which will take you to support.apple.com/en-us/HT201616
EXCEPT June 2020 this link no longer provides any information on this subject. Perhaps Apple no longer allow you to do this. You would have to ask them about that. Or read on, there are other solutions in this FAQ.
This might be because the track is not actually on your computer, maybe it is on a CD or in the cloud. You do have to have the track as a file on your computer before Transcribe! can load it.
On Mac: control-click the track in the "Music" app (or iTunes as it is called in OS 10.14 or earlier) and select "Show in Finder" on the pop-up menu. If this command isn't there or is disabled then the track is not on your hard disk. If it is on your hard disk then a new Finder window will open, showing you the sound file (often an m4a or mp3 or aac file). You can drag-n-drop the track from the Finder window to the Transcribe! window.
If it's not on your hard disk then on the pop-up menu you can use the "Get Info" command instead, and click on the "File" tab to see where the track is. If it is in the cloud then there may be a little cloud icon with an arrow. Click this to download the track.
On Windows: basically the same procedure except you right-click in iTunes and "Show in Windows Explorer".
You have to have the track on your disk in order to load it into Transcribe!, which means that streaming services are a nuisance. You have various options:
Further note about Spotify: There are apps which can access Spotify tracks on phones and tablets but as far as I know it can't be done on a computer. For more info see this page, and this page, and this page. Update summer 2022 : There has been an announcement from Spotify which suggests that even mobile devices will no longer be able to change speed and pitch of Spotify tracks after 1st September 2022.
This is becoming more difficult as the internet giants all want us to sign up to their streaming services instead.
Amazon: sell mp3 files for download, however not necessarily in all countries. Try it and see if it works.
Apple iTunes: You can still buy tracks for download from iTunes, which you can then open in Transcribe!, whether or not you have an Apple Music subscription. These instructions come from this page.
Also note that if you do have an Apple Music subscription then you can click the "..." dots to the right of the track name in Apple Music and the pop-up menu which appears has an item "Show in iTunes Store", where you can buy the track and download it.
This is a useful thing to be able to do, both for the purpose of recording sound while it is playing from a streaming service like YouTube, and also for recording Transcribe!'s output while playing. Of course Transcribe! has the "Export sound file" command, but if you want a file where the speed or key changes in the middle of the track while playing, "Export sound file" cannot do this.
Transcribe! itself cannot record the sound that is playing on your computer's speakers, but there are other programs which can. See this page for a discussion of methods covering Windows, Mac and Linux. You will find plenty of other options too, if you do some appropriate searches.
Stem files are separate soundtracks all belonging to one finished recording, that is separate tracks for the vocal, the drums, the bass, the guitar etc.
On Transcribe! before version 9.3: The best way to use these in Transcribe! is to create or obtain the stems as a single file with multiple tracks, as an mp4 or mogg file. Transcribe! will treat these as a single transcription and will display the multiple tracks in a single window - you will find the relevant commands on Transcribe!'s View menu.
Transcribe! before version 9.3 can of course load multiple sound files in separate windows but does not have any way of loading separate sound files in the same window. However if you have multiple separate stem files then you may find this FAQ (above) "How can I copy a marker layout from one transcription to another?" helpful.
Transcribe! version 9.3 has special handling for stem files where the stems are in separate files. If you are running 9.3 then see the Stem Files command on the File menu, and press the Help button in that dialog.
You can download some example stem files here.
This might happen because the track is in a format which Transcribe! doesn't read. Here are some suggestions.
First, if you have the file on your computer but Transcribe! can't handle the format, then you may be able to make changes to your system to fix this. For example:
If this solution doesn't work then read on...
This is the "Jump List", a pop-up menu showing recent files when you right-click the Transcribe! icon in the Task Bar. There are two possible problems that can happen here. Neither of them are specific to Transcribe!, they can affect any application.
1: Recent files don't appear at all. In this case you may need to "re-establish" the use of Transcribe! for opening xsc files. In File Explorer, right-click an xsc file and select Open With. It should show Transcribe as the default app. Check the box which says "Always use this app to open xsc files" and press OK.
2: Recent files do appear but when you click on one, you get a message "The item you selected is unavailable". This appears to be a long standing Windows bug, not just affecting Transcribe! e.g. see here.
However I tried the workaround mentioned there and it works for me:
- Right-click the icon in the task bar so the menu pops up.
- Do not click the item you want to open, instead right-click it. Another menu pops up. Select "Open" on that menu. It works.
This is crazy. It is Windows (not Transcribe!) that handles jump lists, and Windows clearly knows where the file is, because the "Open" command works. So why does Windows tell you that it is not available? I don't think there is much I can do about this.
Of course you could instead use the Recent Files menu in Transcribe!
You can use the "Open" command (File menu) to open an audio CD track directly in Transcribe!, but this is
not usually the best way. Although your CD player is of course capable of playing in the usual way, it may nevertheless be sluggish
when used with Transcribe! - this is because Transcribe! needs access to the audio samples and so has to use a
different method for reading the data from the CD.
The answer is to use the "Export Sound File" command (File menu) to copy the track to your hard disk, then load that file instead. This gives better performance and also has the advantage that the track will still be there after you take the CD out. There are also many programs other than Transcribe! which can copy ("rip") tracks from audio CD to hard disk.
By the way the "Track 01.cda" etc. files which Windows shows you on an audio CD are not in fact sound files and there is no point in copying these to your hard disk.
Sometimes mp3 files have faulty data in them, and different programs handle this in different ways. Transcribe! chooses to stop, because there is a danger of speaker-destroying or ear-destroying noise. So if you have some other software which plays the whole mp3 file, listen closely at the point in the track where Transcribe! stops : you will very likely hear a strange noise and/or a dropout caused by faulty data. The solution is to update the software you are using to produce mp3 files. Or load the file into some editor which will read the whole thing and re-export it, to get a copy without faulty data.
"Nahimic 2" is a sound effect app used for gaming applications, and I've had a report that it can cause difficulties opening files in other appplications including Transcribe!. I've no idea how common this is.
Here's a basic checklist. If this doesn't solve the problem then please send me the answers to these questions.
1) Is this a new problem? If so, something has changed about your system but what?
2) In Transcribe! - Preferences - Playback, what sound output device is selected? Try a different one.
3) Can other programs output sound?
4) Does Transcribe! appear to think that it is playing, that is, does the playing position indicator move across the waveform display?
5) It might be helpful if I could see your Transcribe! "System Info". After the problem has happened, select the System Info command on Transcribe!'s Help Menu. Press the "Copy" button, then paste the info into an email for me.
People sometimes report all kinds of strange problems using particular sound output devices, and one simple way of bypassing these problems is to install another sound device. If you go to Amazon and search for "usb sound adapter" you will find dozens of little usb sound devices, many of them costing less than $5. Get one of those, plug it in, and use Transcribe!'s Preferences - Playback to select the new sound output device.
This is probably a soundcard or driver issue : it doesn't normally happen if you are using the built-in
sound system, it only happens with third-party sound hardware. In fact this seems to be commonest on the Mac
- I've had it reported for Alesis Multimix8 and Focusrite Saffire & Thunderbolt.
If a soundcard driver crashes then sound output stops, and usually Transcribe!'s play position indicator -
the line moving along the waveform - stops moving.
First, look at Transcribe! Preferences - Playback, uncheck the "Use default device" checkbox, and select the correct output device which you want to use. If you are on Mac and you see "Boom" listed then avoid it - I have had reports of problems with Boom.
Make sure you have the latest version of the software and drivers to go with your sound hardware, then configure it to increase the size of the playback buffer.
Or use Transcribe!'s Preferences - Playback to select a different sound output device.
Transcribe! itself doesn't do this but it shouldn't be a problem, you just need to use other software for playing your instrument through the computer, and have that software running at the same time as Transcribe!. If you search for "play guitar through computer" (without the quotes) then you will find lots of solutions.
(Note: Transcribe! itself does not have ASIO support)
ASIO & RTAS drivers achieve low latency by taking exclusive control of the soundcard so other applications cannot output sound at the same time. There are two possible workarounds. (i) Install another soundcard and configure Transcribe! to use it (in Preferences). If you search Amazon or eBay for "usb sound dongle", you can buy them for two or three dollars. (ii) Don't use ASIO or RTAS, instead tell your VSTi's to use the standard drivers that are part of the OS (meaning that you have to put up with higher latency).
These apps for routing sound on Mac attempt to emulate a real (non virtual) sound output device. I have had varying reports about whether they work with Transcribe!. Transcribe! uses perfectly normal methods for outputting sound and works fine with real output devices. If output is not working with one of these virtual interfaces, it means something is wrong with their emulation and you should ask them about it.
I have no definitive information but I have heard it said that this is a problem with the Realtek drivers used by DELL. Here is a workaround which may sound crazy but try it. Launch Groove Music or Windows Media Player (same thing, different names), drop a sound file onto that player, play a little and then pause it but leave the window open. Now try Transcribe! again. Works for me.
There is also a more radical approach which you may try at your own risk, which is to change the driver. I have heard that Microsoft's "High Definition Audio Device" driver is good but I don't know where you would find this and I haven't tried it.
I don't think this is a Transcribe! issue - Transcribe! uses standard methods for outputting sound on all platforms, and doesn't even know whether there is any bluetooth involved. It is however a common problem for any system and if you search the web for "bluetooth playback delay" then you will find information and solutions.
This could be related to GStreamer versions. GStreamer had a major not-backwards-compatible upgrade from the 0.1 version to the 1.x version in 2013, see the 2013-03-15 entry here. It's ok to have both GStreamer versions installed side by side, but sometimes one or other of them may be absent or broken. Transcribe! 8.6 and later uses GStreamer 1.x, so that's the one you need to install or update.
If your distro has any packages containing "extras" for GStreamer then install those too as that will probably include plugins for things like mp3 which might not otherwise be included.
Sound output problems seem to be particularly common with low latency kernels and the Jack audio drivers. It is related to the sound drivers installed on your Linux distro, and the GStreamer plugins which Transcribe! uses to talk to those drivers. Fixing up your GStreamer installation as described will probably fix the problem but if sound output is still not working then bring up Transcribe!'s Preferences - Playback and you will see where you can specify which GStreamer plugin to use for output. Press the Help button on that page, and follow the instructions there to identify a GStreamer plugin which works when used from the Terminal. Then specify that plugin for Transcribe! and all should be well.
First try playing the video in the default GStreamer-based Linux video player such as Totem, now called "Videos" on Ubuntu 22.
If it can't play it then this is a GStreamer problem. I have had a report that Ubuntu 22 has problems with videos and that you can fix it with
sudo apt remove gstreamer1.0-vaapi
Also see here.
I get occasional reports of this, it seems that Sibelus sometimes insists on exclusive access to the sound card. Here is the solution.
In Sibelius, hit the "Play" tab in the top ribbon. Under "Setup" click the very small diagonal arrow. At the bottom of the next window click "Audio Engine Options". The box "Release device when Sibelius is in the background" must be checked.
Alternatively, see "I'm having a problem outputting sound." above.
I get occasional reports of this, it seems that Finale sometimes insists on exclusive access to the sound card. I don't think it affects all versions of Finale, and also it can depend what version of Windows you are running. I have a report from one user who said that changing Finale's audio setup to use DirectSound, and then making sure to launch Finale before Transcribe!, solved the problem and both programs can then play sound at the same time.
Alternatively, see "I'm having a problem outputting sound." above.
I've had this reported for Band in a Box, I don't know whether it may apply to other programs. The person who reported the problem solved it and told me this: "Turns out the Audiofuse driver has 4 different types of input: cue 1, cue 2, main, and digital. I was trying to use main for Band in Box and Transcribe!. Turns out, for unexplained reasons, that causes Band in a Box to take control of the inputs and Transcribe! was ignored. When I put all inputs on Cue 1, everything works!"
Alternatively, see "I'm having a problem outputting sound." above.
The point here is that the Fx window also uses keystrokes for its own purposes - for instance there
are boxes in the Fx window that you can type numbers into. It would be perfectly possible from a technical point of view for the Fx
window to use the keystrokes it's interested in and pass the rest across to the transcription window in case they
might be shortcuts, but the potential for confusion would be enormous.
The solution is to use the <esc> shortcut which does work in the Fx window, and which takes you straight back to the transcription window.
You can also use the spacebar to get back to the transcription window.
Another solution is to use Transcribe!'s "global shortcuts" - see the Help for Keyboard Shortcuts (File menu).
Transcribe! cannot work miracles, but there are techniques which may help. The first is Karaoke mode, which means
combining the left and right stereo channels but with one of them phase-reversed. This has the effect of cancelling out
anything which is panned in the centre of the mix, which often includes the main vocal. However if the recording is mono,
or if the thing you want to remove is panned to the same place in the mix as things you want to hear, then it won't help.
Also you can listen to either channel on its own if you want.
For more info go to the Mono/Karaoke page of the Audio Effects and Controls (Fx) window, and press Help.
The EQ page of the Fx window may also be useful. However if the thing you want to remove is in the same part of the frequency spectrum as things you want to hear, then once again it won't help.
Don't forget that if the thing you want to hear is very low or high pitched then using the Tuning page of the Fx window to raise or lower its pitch by an octave or two can make it easier to hear. Bass parts are often much easier to hear when raised by an octave or two.
Also please see "Using AI techniques to separate instruments from a mixed track".
You can have any speed you want from 5% (one twentieth speed) to 200% (double speed) by using the
slider on the Speed page of the Audio Effects (Fx) window, or by typing into the box on that page.
If you type into the box then you can use up to three decimal places so you can have a speed such as 96.528% or whatever.
But perhaps you want to set the speed without bringing up the Fx window.
You can choose what speed buttons appear on the toolbar, in Preferences - Toolbar. Starting with version 9.10 you can have whatever speed buttons you want, but in earlier Transcribe! versions you can only select from a list. If you want more preset speeds than that, then you should use keyboard shortcuts (see Help - File menu - Keyboard Shortcuts). There are commands for increasing and decreasing the speed, and you can create new keyboard shortcuts for any speed in 5% increments from 5% to 100%, and then in 10% increments up to 200%.
Also remember that you can adjust the speed using the mousewheel on the speed button in the toolbar (see Help - Toolbar), or using a continuous MIDI controller (see Help - File menu - MIDI Shortcuts).
Also if you want to get seriously technical about it then see Help - Various Topics - Automation.
Transcribe! does not have any magic way of determining the tempo so you must first place markers : set the music
playing and tap the M key at the start of each measure. You don't need to mark the whole piece, maybe about 8 measures
is plenty. If you mark only a couple of measures then the result may be inaccurate if the markers are not very
exactly placed. If you want to measure the tempo in bars (measures) per minute then measure markers are all you need.
If you want to work in beats per minute then you should mark the beats too, so in 4/4 you would tap "MBBB" to mark
a single measure. Or you could mark just the measures, then use auto-subdivision to mark the beats (see the Help
for the Markers menu). Then select the marked part of the piece using the mouse.
If you just want to determine the tempo then use the Compute Tempo command on the Markers menu.
If you want to change it then bring up the Speed page of the Fx window. This page will show you the speed in bpm and you can change it to what you want.
If you want to play the new version on other devices independently of Transcribe! then you should use the Export Sound File command to save the modified sound file. There are some more FAQs about that elsewhere on this page if you need them.
Yes you can, see the "Speed Up" command on the Play menu. Some people who are using Transcribe! as a play-along practice tool like to do this. You can also create more complex practice sequences of loops, speeds etc. by using automation. See Help - Various Topics - Automation, which describes the use of scripts to control Transcribe!. Scripts can do a lot of things, but in the Help there is a specific example for speeding up while looping. You can download this script among others, on this page.
Yes you can, but this does require the use of scripts. Some people who are using Transcribe! as a play-along practice tool like to do this. See Help - Various Topics - Automation, which describes the use of scripts to control Transcribe!. Download the example scripts on this page and try the one called changekeys.xscscpt. You will need to edit it to get the sequence of keys that you want.
For example suppose you have a track recorded at A=438 Hz and you want to adjust it to 440 Hz.
Obviously one way of doing this is to set it playing and adjust the cents setting on the Tuning page of the Fx dialog, until the track matches the tuning of your carefully tuned A=440 instrument. But maybe you want to do it by the arithmetic.
The actual formula for converting a change of frequency to a number of cents is:
cents = 1200 * log-base-2(new_freq / old_freq)
so you need a calculator that can do base 2 logarithms! But maybe your calculator (like mine) can only do base-10 and base-e so let's put it this way, which works regardless of log base:
cents = 1200 * log(new_freq / old_freq) / log(2)
E.g. converting 338Hz to 440Hz (using base 10 logs):
1200 * log(440 / 438) / log(2) = 1200 * 0.001978566 / 0.301029996 = 7.89 cents (so use 8)
In fact for very small changes it's pretty much linear at around 3.95 cents per Hz.
The volume control works by scaling the sample values, which is quite separate from the system volume control. In the centre position (reported as 50 on the Misc page of the Fx window) the samples are passed through unmodified. By increasing the volume setting in Transcribe! you can boost the sample values quite a lot. The reason the gain can be set so high is for dealing with files recorded at very low levels. This is definitely useful.
But if the file you are using is recorded at a high level, so that the waveform almost fills the display vertically, then boosting the samples in Transcribe! cannot have any useful effect as they are already near maximum. In fact it will merely cause Transcribe!'s compressor to kick in, which you can see from the red line appearing below the volume control (this first appeared in version 8.71 a long time ago). So if you want to increase the volume of such a file then you will need to use the system volume control.
You can tell Transcribe to click each time playback passes a marker - see the Misc page of the Fx window, and click the Help button there for more info. You can also use this for generating a metronome count-in by placing markers before the place where you want to start. Unfortunately you can't place markers before the start of the track so this will only work if you don't want to start at the very beginning of the track. We will probably provide a better way of doing count-ins in the future.
Yes you can but we don't provide a fully automatic way of doing this because normally you will need to edit the chord guesses and remove the bad ones. So instead, the way to do it is to copy the chord guesses into Text Blocks, edit them there as necessary, then export the text blocks to a file (this command is on the File menu, or the Text menu in older versions of Transcribe!) so you can print them in any way you want. The best way of copying chord guesses into text blocks is to use the Piano Roll - see the Help for the Piano Roll menu. You can shift-click the piano roll to see chord guesses, and you can right-click (control-click on Mac) to get a menu for inserting a selected chord guess into a text block. Remember that you may need to adjust the track tuning as you will not get any useful guesses if the track is off pitch.
No. Currently my view is that on most material there would be too many mistakes for this to be useful. It's up to you to look at the guesses and decide which ones are in fact notes that you want, and which are not.
You can adjust Transcribe!'s keyboard timbre on the Tuning page of the Fx dialog and you can adjust the volume by moving the mouse up and down while playing a note on Transcribe!'s piano keyboard. The sound used is pretty crude, synthesized internally. It's not intended to sound beautiful, it's only intended as a pitch reference.
I don't plan to turn Transcribe! into a "performance instrument" in this way (for one thing, the built-in sound is not very pretty).
However it shouldn't be necessary - the right way to do it is to use other software to play the sounds from MIDI.
What software are you using now to play sound from MIDI? You should be able to leave it running while you use Transcribe!, and it should continue to work.
It may be a good idea to switch off Transcribe!'s own MIDI responses to prevent confusion. Also if your MIDI player app stops playing when you bring Transcribe! to the foreground then it may be possible to fix this. It's a matter of telling the player app to continue responding to MIDI even when in the background. Otherwise, many apps suppose that if they are in the background, you don't want them to respond to MIDI. Check the Help for your MIDI player app to find out how to do this.
Alternatively you might choose to keep your MIDI player app in the foreground, while using "Global shortcuts" to control Transcribe! in the background. See Transcribe!'s Help for Keyboard Shortcuts (File meu) about global shortcuts.
The short answer is no. The piano keyboard aligns with the spectrum above it, indicating which notes
appear at which point of the spectrum. A piano keyboard is the most suitable way I know of for graphically
indicating a full range of notes from left to right.
However there is help at hand for string instrument players who are wondering where to find a certain note on their instrument. If you right-click (control-click on Mac) a keyboard note then you get a pop-up box showing string and fret information for that note. You can select your particular string instrument and tuning in Preferences - Fret Display.
You should take a screenshot and use that.
If you read music on the guitar then you will know that the open top string is written as E in the top space of the treble clef, but Transcribe!'s spectrum of a recording of a guitar playing the open top string shows it as E above middle C - the bottom line of the treble clef. This is quite correct, because the guitar is a transposing instrument which sounds an octave lower than written : the open top string in fact sounds as the E above middle C although it is written in a guitar part an octave higher. The same issue applies to double bass.
Yes certainly. See Help - Various Topics - Video, and also see this page for more information about video formats and how to use them with Transcribe!.
No, you have to download the video to your hard disk first. See this page for information about downloading videos.
If you are running Transcribe! 8.6 or later: Yes on Windows, Mac, and Linux. On Windows 8.6 - 9.1 and Mac 8.6 - 9.0 you will need to install GStreamer.
Click here for information about GStreamer.
If you are running Transcribe! 8.5 or earlier: Yes on Windows and Mac, but not Linux. On Windows you will need to install QuickTime (it's a free download from Apple), click here for more info.
It's a fun little feature. Of course it's useful for creating videos with speed and pitch alterations, but the time-offset slider in the Video Viewer means that you can also use it as a very quick and simple way of correcting synchronisation errors.
With some videos, when you start playing (or go back to the start of a loop) it can take some time - up to a second - for the video to get synchronised with the audio. One workaround of course, is to start playing a little bit before the section you are interested in. This relates to the way in which the video is encoded and you may be able to improve matters by using a different encoding. See this page for more information about video formats and how to use them with Transcribe!.
First make sure you have the latest version of Transcribe!
If you are running Transcribe! 8.5 or earlier on Windows or Mac then also make sure you have the latest version of QuickTime. On Windows, that would be version 7.7.9, released 7th Jan 2016, click here for more info. Also see "Special note about YouTube videos with Transcribe! 8.4 on Windows 7" on this page.
If you are running Windows 10 or Mac OS 10.10 or later, then you can use Transcribe! 8.6 or later which uses GStreamer for playing video, click here for more info.
If this doesn't fix it then it can sometimes happen that although other software can play a video smoothly, Transcribe! cannot. This is because Transcribe! has to use a different way of playing the video, as it needs to control the playback speed very precisely in order to synchronise the video with the audio at any speed you choose. Some video formats work better than others for this purpose.
These problems relate to the way in which the video is encoded and you will probably be able to improve matters by using a different encoding.
See this page for more information about video formats and how to use them with Transcribe!.
If you can run the latest version of Transcribe! version 9 then first do the update (free to existing users) which
you will find here along with the system requirements. You should also delete the
separate version of GStreamer which you installed, as the latest Transcribe! has GStreamer built-in.
Instructions are here. If you can't run the latest Transcribe then read on.
Look in Transcribe!'s "System Info" (on the Help menu), where you will see a message about GStreamer, either to report what version was found, or possibly to say that it wasn't found. On Windows and Mac, the version which we distribute works pretty reliably, whereas some other versions don't seem to work so well with Transcribe!.
So what you should do is try the version of GStreamer which we distribute here, and make sure to first delete the version you currently have installed, as described on that page. Then restart Transcribe! and check your System Info again.
You probably loaded the sound file again, when you should have loaded the XSC (transcription) file which was saved. Read the Help for the File Menu (in Transcribe!'s Help) for more information.
Set your current transcription the way you like it, then Preferences - New Windows - "Set defaults from current window" and press "Ok".
This applies to things like: the position and size of the transcription window, whether to show the text zone, whether to show the navigation bar, whether to show the spectrum, whether to show the time line, the piano position (top and bottom notes), whether to show video, etc.
You can also choose to set defaults for audio effects if you want, by selecting the appropriate option before pressing "Ok".
This only affects newly loaded sound files - when you load a transcription (xsc) file, you get the settings that were last saved in it.
You can create keyboard shortcuts for the commands QuitNoSaveNoVeto which will close all open transcriptions, or DocWindowCloseNoSaveNoVeto which will close the current transcription. Use "Keyboard Shortcuts" on the File menu, and hit the Help button on that page for details.
On the latest version of Transcribe! (9.25 or later) you can assign these keys. However on earlier versions of Transcribe! some keyboard shortcuts are hardwired already - the ones which are listed against the corresponding menu items when you look at the menu. 't' is New Text Block, 'm' is New Measure Marker, '.' is Stop Playback etc. You can't change these. This only applies to those shortcuts which actually appear alongside the menu item.
We now report and recognise keyboard shortcuts always by the main character which appears on the physical keytop. It should always have been this way really. So if, for example, you had a shortcut created in a previous version of Transcribe! for alt-shift-! then this will not be recognised in version 9.25 or later because that same keystroke is now recognised as alt-shift-1.
To fix this, you should use the keyboard shortcuts dialog, highlight the shortcut in question in the list box, type the desired keystroke in the "type a key" box, and then in the "Change the keystroke" box press the "Use normal shortcut" button.
This replaces the keystroke which triggers the shortcut in question. It's still the same keystroke, just described differently.
I apologise for the inconvenience but the new way is definitely better. It only has to be done once and remember that if you have more than one computer then you can Export keyboard shortcuts from one machine and Import them on another.
For more info, see the 9.25 release notes.
The most likely cause is that when you ran the "MIDI Shortcuts" or "Foot Pedals" dialog,
you configured the shortcuts but left the "Respond to... " selection (at the top of the dialog) as "Never".
Also be aware that if you unplug the MIDI interface or the pedal then Transcribe!, not being able to find it, will set "Respond to..." to "Never" in order to avoid endless error messages as long as the thing is missing. So when you plug it back in again, you will need to run the dialog again to switch responses back on again.
The Magic Mouse can sometimes be a bit over-sensitive to scrolling. To keep it under control, you can try
adjusting it in System Preferences - Mouse.
Also Transcribe! 8.4 or later has an adjustment for mouse wheel sensitivity in Preferences - General, and in fact you can turn it off altogether. Or get a proper mouse with a physical wheel.
This problem affects the Infinity USB-1 and older USB-2 V14 models with macOS 12 Monterey :
for more information read this discussion thread all the way to the end.
However it does seem that more recent USB-2 V15 model works ok, and the current model - the Infinity USB-3 - also works ok with macOS Monterey.
Apparently this is caused by an incompatibility between the chip in those older pedals, and macOS 12. This causes the Mac to mistake the pedal for a mouse. Hopefully Apple will fix it but until they do, you can prevent it happening by positioning the mouse over some part of the Transcribe! window which does not respond to mouse buttons, such as the right hand side of the lower toolbar (not on any control).
Strange effects can also happen if you have not in fact selected the pedal in Transcribe!'s Foot Pedal dialog. I find that on macOS 12 the foot pedal selection may default to "Headset HID device" whatever that is. Transcribe! will say that it is "working properly" but it's the wrong device so use the pop-up menu to select the proper pedal device. In Transcribe! 9.2 that dialog has been updated so it always reports which device is "working properly".
Transcribe! 8.4 and earlier for Windows uses QuickTime (if installed) for video and other purposes (and does not use GStreamer).
This is a free download from Apple. It's very useful with Transcribe! because it allows Transcribe! to display videos, to read various sound file formats not otherwise readable, and to export compressed M4A sound files (On the Mac, QuickTime is already included). The link is http://support.apple.com/downloads/#quicktime However for some unknown reason this page of Apple's sometimes fails to list the latest version of QuickTime for Windows which is 7.7.9, released 7th Jan 2016, so you can also download it by clicking here.
NOTE: when installing the latest version of QT you will need to select "Custom install", click on "Optional QuickTime Features" and select "Entire feature will be installed...". This is needed for Transcribe! to be able to communicate with QT.
You will need to reboot the computer after installing QuickTime, before Transcribe! will be able to find it. If you look in System Info on Transcribe!'s Help menu, then you will see a message telling you whether QuickTime was found, and what version.
Apple have announced (April 2016) that they are no longer updating QuickTime for Windows. The last-ever version of QuickTime for Windows is version 7.7.9, released 7th Jan 2016. This means that there are a couple of security vulnerabilities in QuickTime which will never be fixed (see
Both of these weaknesses would only be an issue if you download malicious content and open it in QuickTime. Reading the advisories, my impression is also that they would only affect mov files though I'm not an authority. Just to put this in perspective, this means that you can get into trouble if you download mov files from malicious hackers. But then, you can get into trouble if you download any kind of file from malicious hackers, because QuickTime is far from being the only software with security vulnerabilities.
Transcribe! up to version 8.5 uses QuickTime (if it is installed) for reading some sound file formats (such as m4a) and for displaying video. If you have installed QuickTime on Windows and now choose to uninstall it then Transcribe! will no longer be able to read some sound files (so you would have to convert them to some other format to use them with Transcribe!) and will no longer display video.
The best solution is to update to Transcribe! 8.6 or later, which does not use QuickTime at all, either on Windows or Mac.
Because Apple are discontinuing the QuickTime API. That doesn't mean that QuickTime files (mov files) will stop being supported, and Apple might well continue to call their bare-bones media player "QuickTime Player", it just means that under the hood Apple are using different technologies to handle these files. GStreamer can of course read QuickTime (mov) files.
I've had a tiny number of reports of this and it seems to be caused by a conflict between the GStreamer version included as part of Transcribe! 9.10, and a previously installed GStreamer version which was used by earlier versions of Transcribe. In the great majority of cases it's not a problem to have both versions and I don't yet know why it is a problem for a very small number of people.
The solution is simple - delete the previous version of GStreamer, which Transcribe 9.10 does not need because it has its own copy bundled in the app. The folder to delete is /Library/Frameworks/GStreamer.framework (on your System Disk - note this is not the same as the Library subfolder you will find in your Home folder - you won't find GStreamer there. To find the Library folder on your system disk: select "Go to folder..." on Finder's "Go" menu, and type "/Library".)
This happens if you have installed the "N" or "KN" versions of Windows, which for legal reasons have various files missing which are needed for handling audio media. You can solve this by installing the "Media Feature Pack"
provided by Microsoft here.
Click here for more information.
Note 1: Micrososft's media player "Groove Music" is included in this feature pack, so after installation of the media pack, try running Groove Music to make sure it is installed ok.
Note 2: An alternative solution would be to do a fresh install of the standard version (not the "N" version) of Windows.
You can transfer transcriptions from one computer to another, even when the computers are
running different OS's (Windows / Mac / Linux), as long as the destination computer supports
reading the sound file format involved.
First read Transcribe!'s Help for the File menu, to make sure you understand about transcription files (xsc) and sound files (wav, mp3 etc).
Then copy both the sound file and the transcription file to the same folder on the destination computer. Or put them in different folders, but then you will probably need to use the "Import Sound File" command to find the sound file on the destination computer.
When you open the transcription file on the destination computer then it can happen that although the xsc file does open (its name is displayed in the titlebar of the transcription window), it reports that it cannot find the sound file. In this case you should make sure that the sound file is present on the destination computer, then use the "Import Sound File" command to load it into the transcription.
You can always find the sound file used by a transcription, by bringing up the Sound File Info window (on the Window menu).
There is no automatic way of doing this for the Preferences. However you can transfer Keyboard Shortcuts, MIDI Shortcuts, Foot Pedal shortcuts, and EQ presets, by using the Export and Import commands in the relevant dialogs.
If you are using Transcribe! with the Video Viewer or with another app such as a music notation program then you may want to reduce the amount of space used by the main Transcribe! window. Here are some suggestions.
The green button in the title bar will switch between full screen and normal window size. However it can happen that the normal window size somehow gets bigger than full screen size, causing this problem (this doesn't just apply to Transcribe!). You can't make the window smaller by dragging the lower edge up, because the lower edge is hidden by the dock. So drag the top edge down, then drag the whole window up.
In version 8.71 we have added a "Reposition this Window" command on the Window menu, to solve this problem.
The "Recent Files" list shows recent transcription (xsc) files. So if you don't save your transcription
(the "Save" command on the File menu) then there is no transcription file, so it won't appear on the
"Recent Files" list. Read the Help for the File Menu (in Transcribe!'s Help) for more information.
The reason we don't list sound files is that if you have saved a transcription file then it is usually a mistake to load the sound file again, and it confuses people. Instead you should load the transcription file because that is where markers, loops, etc. are stored.
The "Reload Last Transcription" preference loads the most recent transcription on the Recent Files submenu, so you must have saved your transcription for this to work.
When you want to load a sound file (e.g. to start a new transcription) then it's very easy to use drag-n-drop.
Transcription (xsc) files store information about your transcription such as
tuning changes you have made, markers you have placed, EQ settings, etc, and of
course the name of the sound file. Transcription files are very small, they do
not contain a copy of the audio data.
If Transcribe! then offered to open the sound file again then you would be starting a new transcription and the markers etc. would all be gone - this would be very confusing for a beginner.
But is is very easy to open sound files directly if you want to. Bring your file browser (Windows Explorer or Finder) to the front (but make sure you can see some of Transcribe!'s window behind it). Locate and select the sound file (or files) in the browser and drag-n-drop onto the Transcribe! window. You can also drag-n-drop from iTunes or (on Mac) Spotlight.
I'm glad you asked - yes there are, on this page.
Remote Buddy enables you to control apps on your Mac using various remote devices including for instance your smartphone. When you are setting up Remote Buddy on the Mac you need to know the application identifier of the target application, which is com.seventhstring.transcribe
Try running this command in your terminal before launching Transcribe! (in the same terminal):
and possibly this too:
Then launch Transcribe! with a line like this (but with your user name of course):