I feel your pain!
In the video editor Final Cut Pro ( not a DAW…), I can send a dialog recroding (eg podcast ) directly to the transcription service Simon Says, edit the transcription using text-editing in Simon Says, then send the edit-data (XML) back to Final Cut and have it assemble (effectively auto conform ) the edited dialog on a timeline. Now it’s ready for fine editing. Works across multi tracks too.
Admittedly there’s a couple of tricks to this workflow … but it works!
Seems to me a DAW should be able to manage this type of thing within the app these days!
3 Likes
yeah an editor has mentioned this to me. Yeah seems like a DAW should take advantage of this given that its a pro audio environment
Final Cut Pro now includes native support for Transcribe to Captions, enabling users to automatically generate subtitles from audio for improved accessibility and workflow efficiency. DaVinci Resolve also features built-in speech-to-text functionality, allowing for automatic transcription and text-based video editing directly within the timeline.
Avid Media Composer has recently enhanced its Transcript Tool, now supporting transcription on both the Source and Record sides of the sequence. This update significantly improves text-based editing workflows, making it easier to navigate and edit content using transcriptions. Although not new, the integration of PhraseFind allows users to search for specific words or phrases within their media assets, further streamlining the editing process. So most NLEs come already equipped with all the tools.
You will anyway be given an edit locked timeline ready for ADR into various languages.
Wondering about the Markers that are read by Wwise in the example. Are these wav markers or specific markers for Wwise? I would love to be able to embed markers into my sound effects before (exporting and import) metadata editing in soundminer.
Marker search is possible.
You can make a simple PLE setup. Here’s a gif with some examples. It is definitely not a perfect solution and has its limits, but it does work.
4 Likes
Thank you, I’ll take a look at that process.
Yes, not perfect since it most likely would not search and sync word-by-word on the timeline.
Well obviously not if the markers contain sentences like in my example.
But that is a question of how the AI Transcription outputs the text to marker…
But honestly @stingray i feel this is a little bit nitpicking and not helpful.
I’m presenting a possible solution and admit it is not perfect… have you tried using it? How did it work?
Instead of just pointing at the a hypothetical flaw, we could work together and provide input for SB to make a solution that fits our need.
Let’s be constructive 
Exactly. We’ll have to discuss this after N14 is released.
No nitpicking intended and not trying to score points. IMHO the devil is in the detail. Unfortunately to provide input you do indeed often have to be hypothetical. I’m aware that your solution (which is what I already suggested in my first post on this above) was not intended as a perfect solution. And, by the way, I fully appreciate your constructive input and was not attempting to negate it. @MattiasNYC was suggesting word-by-word sync and search (although I may have misinterpreted this). The ability of a PLE solution to provide that would be dependent upon the transcription AI. As I also suggested above, SpectraLayers’ transcription AI already does something close to word-by-word sync but you can’t move the play cursor according to a word search (AFAIK). If Nuendo could have this then I believe it would be extremely useful to a lot of users.
I’d already used similar PLE name search methods prior to your post and, as mentioned, posted the same (or similar) solution a few posts above yours. And yes, PLE can work very successfully for this.
FWIW I consider this commentary and my previous commentary to be constructive.
1 Like
I don’t suppose that they FINALLY made it possible to render in place a MIDI file in mono. How long have we been asking for that?
Audio segment editing is great! Would be great if it could be done with just a mouse scroll after highlighting the range.
You can add the Translated text to any Marker attribute of your choice.
Fredo
Yeah i understood that from the video. But the thing is, will it output markers per word OR per sentence? This matters for finding alt takes etc. However you would get hundreds to thousands of markers on a film project. How does Nuendo handle that? I’ve seen it struggle with too much markers.. zooming and navigating becoming very slow (at least on my m1max).
I’ve already worked with macwhisper (not imported as csv just in a spreadsheet) and found it had issues with finding comma’s points and words in non-english speech. In English it fared pretty well but sentences where hard and special words and names get botched up easily. We’ll have to wait and see how it all works, but it is exciting for sure.
The feature is designed for ADR. Which means that you already have set your markers for each line/sentence you want transcripted.
You can also create new markers and let the algo create a marker per sentence. That however is far from accurate, but good enough to transcipt longer parts.
HTH
Fredo
1 Like
Understood, now imagine you are a dialogue editor on a documentary.. you need an alternate line or a word for a scene that is missing that word (offf axis, noisy, distorted, whatever).. since this is docu it is unscripted. You can either listen to all takes or use AI.
Or you are working on a podcast, you have a lot of alt takes from a long session.
Or you have day of vehicles recordings with a lot of slates that describe the session details.
As you can see: There are many more situations in post audio then ADR where AI transcription can be very useful. Why not service more of them by making the system flexible for the user. Customisation is why i choose Nuendo over other daws (years ago) let’s keep that an integral part of Nuendo for the future!
And that is exactly what will be worked on in future. The N14 Speech-to-text is just the basis and the beginning for a whole range of new features and additions, that will come to Nuendo in future releases.
11 Likes
Descript Style Text based editing is on the horizon. Nice!
Now we need a simple feature which detects or calculates the average BPM of a selected audio event without altering the tempo track or overall project tempo. Oh wait… this feature was a part of an old Beat Calculator until it was depricated a year ago in Cubendo 13.
Bring it back then.
4 Likes
Thank you, thank you, thank you!!! 
Should be interesting to see the upgrade price from Nuendo 13 to 14. If the upgrade price from 12 to 13 sits at £132 with a free upgrade to 14, then I am hoping 13->14 is in double figures!?
It will probably be the same price.
AFAIK upgrade prices don’t go down with time, so just because 14 will be the most recent release then, 13 is the current most recent release, so it won’t make a difference. Save for a plain old price increase, of course.
The free upgrade to 14 for those who upgrade to 13 now is just what the call a “grace period”. That’s common practice where if you upgrade to the current version within a certain time window from a new release, you get the newer version when it comes. They do that so people spend their money at once, instead of holding it until the new version is available.