Any feature requests? Let us know by posting them down below!
A “Voiceflow Sound Library”
This would be complementary to the official Alexa Sound Library. The problem is that we could use a lot more sounds, and many VF users might share the same need for the same sounds. For example, “Duck Quack” is not in the Amazon sound library. I have to find or make my own. It takes time to find the sound, verify that it is free for commercial use, and possibly edit it (and maybe convert it if I host it myself). This takes time. But then many other VF developers might eventually have a similar need for a duck quack sound. This would be a nice benefit that VF can offer Devs.
These would be mostly short sounds, one to four seconds or so. Animal noises, car noises, fighting noises, etc.
The VF sound library would have categories similar to Amazon’s and maybe more refined categories as well. For example, Animals/Mammals, Animals/Birds, SciFi/Lasers, SciFi/Spaceships, etc. VF would have a set of web pages and host the sound clips. VF Devs would browse for sounds, and grab the URL. The set of web pages in the VF website would include being able to listen to the sound clip, similar to the Amazon one.
Bonus: Have a way to access the sound library from the project canvas, and bring in a sound into a speak or combine block without having to copy and paste the URL. This might be a better approach if you want to “hide” the public URL (or use some other method to link them), so that non VF outsiders don’t try to hot-link to them, or inappropriately download them for other non-VF projects.
There might be several ways to go about obtaining the sounds.
The first would be to purchase or lease a sound library from a third party provider. Something like this will probably be expensive, and as such maybe use of the VF library might need to be limited to paid plans. Or, have a smaller set of basic sounds for those subscribed to the free community plan, and a larger set available to those on paid plans. Or, charge a small per-use fee (a dollar per clip, per skill, or something like this). For VF, there would be the inherent cost of hosting the audio clips, maintaining the library website, and the purchase or licensing of the sound library from the third party provider.
Over time, add more sounds and expand the library.
Another way would be for members to contribute sound clips. However a way to assure that the sounds are free for commercial use, not copyrighted, etc, would need to be established, and would require resources for verifying the sounds can be used without risk of IP infringement.
Also whatever the source of the sounds, they may need to be mastered for appropriate loudness for use within an Alexa skill, as well as converted for use as in-line audio.
This is a great idea We’ve thought of similar use cases once we incorporate the marketplace into Voiceflow eventually …
Thanks for sharing
Great idea Mark!
Why not a template library ? Where sharing project or ideas between VF users .
Like a private beta:-)
Hi guys, of couple of issues.
- When Google Support comes outs and Id like support for the SSML Google Tags such as par & media
- Notifications support as permissions for Google and Alexa, please. (Aim to design examples use cases)
- Can we have a control Z to undo text changes? I just deleted my copy in a Speak Block and could undo my mistake!
- Comment Block. Can only drag and move it from the bottom. Could this be improved?
- Not being able to work between two different tabs (two different projects is a little annoying and slows me down when I have copy objects between projects.
Thanks as always, Ill keep adding as I find them.
HI, my one of my api’s takes a few seconds to load (month forecast) and i have been trying to add a block before the api that could say something like, loading, please wait. The block works but only after the same delay that it takes to get the forecast, example…
Choice > (delay) Forecast > etc…
Choice > (delay) Speak > Forecast > etc…
Choice > (delay) Code > Speak > Speak > Forecast >etc…
Thats a very crude example but im sure you understood with it anyway lol so, no matter what i put in front of the api block, the dl time will always be in front of them !
After many trial and errors, all ended in failure lol
Spoken to Nicolas and has told me that i need Progressive Response to be able to notify the end user prior to calling the api. Yours hopefully
Just a minor thing… longer variable names would be very helpful.
For sure! We’re looking at adding this soon
Hi folks, how about .wmv format for the stream block please?
wmv is a “old” video format, not very used anymore and not compatible with Alexa or Google. But you can use online converter to extract the audio track to a mp3 file.
It’s video from my weather camera
So if you want the video you need a display block not a stream block no? If your webcam can not export/stream other format you will need to use an API to convert your wmv to a mp4 format or a server script (or a software if you run it locally). Another solution is to use static images (often jpeg) from you webcam and load them in an array then pass this to the APL. @Bill made a tutorial that can help you on this : https://hearingvoices.ca/2019/02/15/alexa-apl-background-image-tester-intermediate/
Add a “no duplicates” feature to the Speak block. It would go hand-in-hand with the randomization feature. Currently, this can be easily achieved with using the random block, and linking each random block output node to individual speak blocks. But it would be more convenient and less cluttered on the canvas if we could load all the replies in a single Speak block and be able to click check boxes for both random and “no duplicates”. This is not high priority as there is a relatively easy workaround, but it would be nice to have. Thanks.
Add access to VF University and the Facebook community from the Dashboard. I often want to look up something in VF University from the dashboard.
Also I see you have access to the VF Forum from both project canvases and the dashboard, but the icons are different.
Not high priority, but this would add some consistency and make navigating to different aspects of VF easier.
Add a new merge block which would enable combining Alexa spoken text with background music or sound effects in an audio file. For example, Alexa is reciting a horoscope reading (which includes the user’s name in a variable), while music plays in the background. It would be a like a speak block, but with two inputs (text and an mp3), which are mixed and played in tandem when the skill plays.
There is an open request for this on the Amazon side, and who knows when it will be done (if ever). Wondering if it is possible to create this on the VF side of things, and have skills be able to utilize it.
The stream block can handle AAC and MP4 so i am going to assume that that it is the block thats used for video? I had a look at Bills tutorial, it does not look like its what i could use, although i do have all the images that are used to create the video but then that would mean having to upload all the images and to be honest, that seams a bit backwards when i already have the video(s). Whilst i do have the option in the software to change it to MP4 the problem is all of my day videos, nearly 1000, are in wmv format, changing the format now would mean many, many days of work converting them to MP4, not to mention all the scripts on my various sites would need to edited to suit this change.
Yes @Bashy, I know it’s some more work here but you will have to do it anyway as wmv format is no more used in new web technology. You can convert your old videos to mp4 automatically with free softwares (handbreak is one of them). Just drag and drop your videos and click convert, simple as that
Definitely need to use the Display block if you need the video (search the Facebook group for APL video, I’ve made some post talking of using the videos with APL). I haven’t test the stream (audio only) block with mp4 for the audio but maybe it can handle it (you’ll need to convert your wmv to mp4 or mp3 if you only need the audio of the video).