Comments

RogerS wrote on 6/14/2023, 8:04 PM

What does that mean automatic? We had speech to text before Resolve and other programs.

Former user wrote on 6/14/2023, 8:34 PM

When this was discussed earlier, it sounded like they planned on bringing the speech to text from Vegas365(Subscription version) to normal Vegas's but also they were planning on charging for it, so most likely you pay for Vegas, then you pay a monthly subscription just for the subtitling.

Anyone opposed to subscription models aren't going to like that approach. I guess if they added in Transcription editing where you rough edit an interview(as am example) via text, you search for where the subject says something important such as "Mary had a little lamb" , highlight text drag to timeline, and that section of audio/video is added to timeline.

At least you would get a whole new way of editing videos with that subscription cost, but I think speech to text should not cost extra, is not an additional cost with Resolve studio, Capcut and Premiere. Problem is they are cutting the fat from Vegas to make it leaner to sell it cheaper, I don't see them absorbing the server cost. The other method that Resolve and Premiere use is to utilize large language models that could become part of the Vegas Deep Learning models, so the processing is done locally on a users GPU or CPU and Magix doesn't have to pay for 3rd party server speech to text services.

RogerS wrote on 6/14/2023, 8:44 PM

I thought Premiere was doing it locally with (not so great) Mozilla libraries. No idea about Resolve as it's recent. Having done local speech processing on CPU it is very resource intensive but if you have processing power to spare it's viable.

There hasn't been an update about pay-as-you-go plans to add online services like cloud computing and storage to non-365 VEGAS, so no idea if that's going to happen.

Nath.veg wrote on 6/15/2023, 8:21 AM

I'm not talking about voice synthesis, I'm talking about automatic subtitles. For example, I import a video with a voice, and the subtitles appear directly on the screen, with a choice to customize the font, etc., just like on Premiere Pro.

Je pensais que Premiere le faisait localement avec (pas si génial) les bibliothèques Mozilla. Aucune idée de Resolve car c'est récent. Après avoir effectué un traitement local de la parole sur le processeur, cela nécessite beaucoup de ressources, mais si vous avez de la puissance de traitement, c'est viable.

Il n'y a pas eu de mise à jour sur les plans de paiement à l'utilisation pour ajouter des services en ligne comme le cloud computing et le stockage à non-365 VEGAS, donc aucune idée si cela va se produire.

 

RogerS wrote on 6/15/2023, 8:26 AM

Already exists in VEGAS and is what we're talking about- "speech to text."

Nath.veg wrote on 6/15/2023, 8:28 AM

Mais cela ne fait qu'ajouter une voix robotique au texte que nous mettons, n'est-ce pas ?

Dexcon wrote on 6/15/2023, 8:34 AM

Just curious ... if it works so well for you in Premiere Pro, why not continue using that NLE for that purpose? Each NLE has its own advantages and disadvantages and no NLE is the be-all-and-end-all of NLEs. Clearly, what you are seeking is not going to happen with Vegas Pro in the next few weeks or months.

Cameras: Sony FDR-AX100E; GoPro Hero 11 Black Creator Edition

Installed: Vegas Pro 15, 16, 17, 18, 19, 20, 21 & 22, HitFilm Pro 2021.3, DaVinci Resolve Studio 19.0.3, BCC 2025, Mocha Pro 2025.0, NBFX TotalFX 7, Neat NR, DVD Architect 6.0, MAGIX Travel Maps, Sound Forge Pro 16, SpectraLayers Pro 11, iZotope RX11 Advanced and many other iZ plugins, Vegasaur 4.0

Windows 11

Dell Alienware Aurora 11:

10th Gen Intel i9 10900KF - 10 cores (20 threads) - 3.7 to 5.3 GHz

NVIDIA GeForce RTX 2080 SUPER 8GB GDDR6 - liquid cooled

64GB RAM - Dual Channel HyperX FURY DDR4 XMP at 3200MHz

C drive: 2TB Samsung 990 PCIe 4.0 NVMe M.2 PCIe SSD

D: drive: 4TB Samsung 870 SATA SSD (used for media for editing current projects)

E: drive: 2TB Samsung 870 SATA SSD

F: drive: 6TB WD 7200 rpm Black HDD 3.5"

Dell Ultrasharp 32" 4K Color Calibrated Monitor

 

LAPTOP:

Dell Inspiron 5310 EVO 13.3"

i5-11320H CPU

C Drive: 1TB Corsair Gen4 NVMe M.2 2230 SSD (upgraded from the original 500 GB SSD)

Monitor is 2560 x 1600 @ 60 Hz

Nath.veg wrote on 6/15/2023, 8:37 AM

I don't use première pro, I just think it's a shame that vegas don't do anything to update, and are behind in a lot of things compared to their competitors, if they carry on like this they're going to lose us all that's all.

RogerS wrote on 6/15/2023, 8:50 AM

Non, ce n'est pas ça.

Speech to text (transcribe text from audio) is not text to speech (une voix robotique)!!!

VEGAS does have it and is not behind Premiere on this:

Last changed by RogerS on 6/15/2023, 8:56 AM, changed a total of 2 times.

Custom PC (2022) Intel i5-13600K with UHD 770 iGPU with latest driver, MSI z690 Tomahawk motherboard, 64GB Corsair DDR5 5200 ram, NVIDIA 2080 Super (8GB) with latest studio driver, 2TB Hynix P41 SSD and 2TB Samsung 980 Pro cache drive, Windows 11 Pro 64 bit https://pcpartpicker.com/b/rZ9NnQ

ASUS Zenbook Pro 14 Intel i9-13900H with Intel graphics iGPU with latest ASUS driver, NVIDIA 4060 (8GB) with latest studio driver, 48GB system ram, Windows 11 Home, 1TB Samsung SSD.

VEGAS Pro 21.208
VEGAS Pro 22.239

Try the
VEGAS 4K "sample project" benchmark (works with VP 16+): https://forms.gle/ypyrrbUghEiaf2aC7
VEGAS Pro 20 "Ad" benchmark (works with VP 20+): https://forms.gle/eErJTR87K2bbJc4Q7

Nath.veg wrote on 6/15/2023, 8:58 AM

Ohhhh okk insaneee, mais je ne peux pas utiliser si je n'ai pas 365 ??

RogerS wrote on 6/15/2023, 9:34 AM

Oui, maintenant c'est necessaire d'avoir 365 pour utilizer tout les services en ligne.

Former user wrote on 6/15/2023, 6:51 PM

I thought Premiere was doing it locally with (not so great) Mozilla libraries. No idea about Resolve as it's recent. Having done local speech processing on CPU it is very resource intensive but if you have processing power to spare it's viable.

 

Premiere started out with Cloud processing the speech like Vegas is doing now, they moved to doing it locally, download the language models you need. I have looked at task manager when Premiere is transcribing, not using much CPU or GPU, yet it's reasonably fast, but it's not so great for accuracy, Being able to separate speakers is nice.

Resolve acts more like the GPU based whisper transcribers, using most of your available GPU and can be very fast, seen transcribe processing rates up to 42x, but also as low as 6x. The good thing about these built in transcribers is the editing via text(transcript) that's obviously a feature you don't get with standalone whisper transcribers and something Vegas could add.