Please note: Microphone control output format for mobile is different and depends upon device used. For Canvas App Microphone control, the audio formats are 3gp for Android, AAC for iOS, webm for web browsers. You will need to check the format before sending it to Whisper and convert if needed (e.g. coming from mobile device?) You could use CloudConvert connector to convert audio file type - learn.microsoft.com/en-us/connectors/cloudconvert/
Hahaha, the only reason I wanted to use Whisper over Azure was voice supporting format😮. I am already using azure and cloudconvert. 😂 Thanks for your note Reza!
Awesome video! I followed very close, but the response from the OnSelect results in a blank .text variable. Looking at PowerApps Monitor, the request "body" shows `filename=\"file_5dcd1b8d-827a-4c8b-bb45-57d58250e96b.webm`. Note the file extension. I'm using Chrome, also tried in Edge, still .webm. Firefox saves audio in .ogg, so that doesn't work either. Which browser are you using?
@@RezaDorrani thanks! Your guide is so much easier than the other guy Ashiq...but how do you upload an audio recording and then convert it into a text document please ? that would be so useful !
Even before watching this video I need to thank you for taking time creating this video back to my request 4 weeks ago. And I should admit that it is more than I was expecting. . Much appreciated!
I didn't think this was possible (despite trying all last weekend) due to the problems with sending multipart forms in a custom connector. This is such a powerful upgrade to my understanding Reza, thankyou. The GPT use cases are huge, and just in time for my talk to the Scottish Power Platform Usergroup tomorrow. Thanks again Reza, your content is always excellent.
As usually this video and your explanation is awesome. Thank you very much for sharing such a nice use case related post. No words to praise your explanation...
Thank you for such contents. I learned alot from your videos. Created a project also, but recently got know about multiplexing. Could you create a video explaining it.😅
Hi, I am a beginner in PowerApps. I am trying the Whisper (Speech to Text) connector by following your video. Please may you tell where to get the Bearer API while connecting the connector in PowerApps. Thanks
Reza, this is amazing, thank you yet again! Would you be able to explain how to create a Power Automate flow that sends an audio file uploaded to SharePoint to Whisper for transcription?
Welcome! I do not have a video reference on it and something that’s not in my planned list of videos. I make videos when I receive multiple requests on a topic. Every 6 months I ask subscribers for topic suggestions on Community Tab of my channel. Most voted idea gets added to my backlog. Make sure to post this idea whenever I post the next topic suggestion post.
I have covered that in other videos related to GPT and Power Platform hence did not want to repeat the steps. Check my playlist on GPT and Power Platform.
Hi thank you again for this video. You didn't explain how you did o the process to convert "attach audio file" to text and the process to convert audio to image. Can you share it with us o did you already make video on those two process?
The attach audio file option uses attachments control. You can simply grab the value from that control and provide it to the API call. I have done videos on attachments control. You can see how to get value from there. Converting from text to document after that is same as how it is explained in video.
Depends on how you have created the custom connector. I will have to look at your App in action to provide guidance. I recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com
Thank you very much, Sir. I'm hoping you'll teach us how to use this API to create, search, filter, and email items in SharePoint lists and Dataverse. This is to ensure that our Apps are also accessible to our impaired and blind clients.
Thanks. I make videos when I receive multiple requests on a topic. On Community Tab of my channel, Im currently running a topic suggestion option. Folks suggestion ideas and most voted idea gets added to my backlog. You can try posting this idea there.
Reza 🎉 this will be so useful for our clinicians. Thanks so much!!! I only have one question once I create the connection to the api can I used it as a component in a web portal using other language or it has to be used only in the power platform?
Most welcome! Video is related to Power Platform. You can use it in Power Apps and flows or call it from bots. Power Apps can be added to Microsoft Search, SharePoint, Teams etc.
Hi Reza, at 3:29 when I add my custom connector, I am not prompted for an API Key. Hence, when I run the app, I encounter 'invalid_api_key error. How to resolve this? I'm using Edge browser.
I have not come across the mentioned issue and hence not sure what the fix for your issue is. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com
I have not come across this issue & hence not sure what the resolution is. I recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com
Hi Reza, i can't find attachment control if not in forms so i didn't quite understand how to convert audio to text from attachment, it would be very useful! Thanks!
Thanks Reza. Video is very helpful. I am getting error on content type. Where did can I get the content type? Thank you so much for your help in advance.
I have not come across this issue and hence not sure what is the cause for your issue. I recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com
Hi Reza, its awesome video. I converted speech into text using whisper. Thanks for that. But I couldn't find AI Builder action 'create text with GPT on Azure Open AI Service'. As it is preview, its not available for my tenant or any configurations required?
Many thanks , Reza Really appreciate all your efforts! I have a question from cost prospective I have E5 license is it enough for ai models or there’s extra cost also for openAi API do you have any budgetary figures to use it within organization? Many thanks again and again
learn.microsoft.com/en-us/ai-builder/administer-licensing You will need a premium license for power apps + you would also need to take into account OpenAI costs related to API calls to their service.
Hi Reza, This is very helpful, thanks for posting. I don't see AI Builder for Create text using GPT. Seems it is only available as a preview feature in certain locations. Is it true? If so, any idea which region it is available and if there is any process involved in getting access to it (other than the Azure OpenAI, I already have subscription and Azure OpenAI Access). Thanks.
Awesome Video, Reza. Right what I was looking for 🙏🏻 One question has occured: As I am working in MacOS/IOS there seems to be a problem with the audio format while recording speech to text. Is there any solution with wbmp-audio-format in MacOS environments? Thx for your help.
I think I showed it in these videos ruclips.net/video/F3OpKuy8wB8/видео.htmlsi=hc4efM-P5BgQ4AlC ruclips.net/video/mzZLkf8pz9U/видео.htmlsi=kcFaYjYiWbJdSQnr
@@priyank1818 I make videos when I receive multiple requests on a topic. Every quarter I ask subscribers for topic suggestions on Community Tab of my channel. Most voted idea gets added to my backlog. Make sure to post this idea whenever I post the next topic suggestion post.
Thanks! I make videos when I receive multiple requests on a topic. Every 6 months I ask subscribers for topic suggestions on Community Tab of my channel. Most voted idea gets added to my backlog. Make sure to post this idea whenever I post the next topic suggestion post.
Thanks Reza, super cool Video! I never thought working with OpenAI was that easy. For longer audios, I am very interested in your example with the audio upload using the attachment control. I am having trouble in building a JSON of my uploaded audio. Do you have a video specifically on that example?
I found a solution, save the attachment to an image control using First(DataCardValue2_1.Attachments).Value. Then use the image control in the JSON function as Set(tempJSON, JSON(Image2.Image,JSONFormat.IncludeBinaryData)) If anyone else had the same issue. Anyway great content Reza :)
@@paolodeluise sorry for my late answer, i was on vacation ;) I dont know if this is the most elegant solution, but i used an image control "img_Rec_Uploaded" to save the content of the uploaded audio. 1. At the image control add this: Image = First(att_Rec_Uploaded.Attachments).Value 2. At the attachment control add this: OnAddFile = Set(tempJSONUploaded; JSON(img_Rec_Uploaded.Image;JSONFormat.IncludeBinaryData)); Set(StringB64VarUploaded; Left(Last(Split(tempJSONUploaded;",")).Value;Len(Last(Split(tempJSONUploaded;",")).Value)-1)) 3. Upload the StringB64VarUploaded variable to PowerAutomate: AIToEmail.Run(StringB64VarUploaded; First(Split(Text(aud_Rec_Uploaded.Duration); ".")).Value; "Text";JSON(ComboBox1.SelectedItems); User().FullName) I hope this helps, good luck!
I do not have plans for a follow up detailed video unless many folks request for it. Every quarter I ask subscribers for topic suggestions on Community Tab of my channel. Most voted idea gets added to my backlog. Make sure to post this idea whenever I post the next topic suggestion post.
Hi Reza, thanks for this turorial. I have one question tho... One client asked me to integrate this AI component with its SQL on-premise database to answer question regarding KPIs or filter information of tables. Do you think this is possible? Thanks!
Hi Reza, Its an awesome demo!! I tried this and when i use it it generates the result in the Textlabel in Hindi, is there a way i can always set the labeltext language as English? Please let me know!!
I do not have a video reference on this scenario and would have to try it out to provide guidance. I recommend checking or posting your query on forums in case someone has done something similar powerusers.microsoft.com
Thank you Reza. If I record the audio on an iPhone, I get an error because the file format is WBMP but the iPhone saves audio in another format. How do you solve this?
Format for mobile is different and depends upon device used. For Canvas App Microphone control, the audio formats are 3gp for Android, AAC for iOS, webm for web browsers. You will need to check the format before sending it to Whisper and convert if needed (e.g. coming from mobile device?) You could use CloudConvert connector to convert audio file type - learn.microsoft.com/en-us/connectors/cloudconvert/
@@RezaDorrani you are right, the file format of the recording on iOS is AAC and the whisper api doesn't support it. Using CloudConvert means that the audio recorded is sent to a third party in addition to OpenAI and Microsoft and that could be a privacy concern. Do you have any other idea?
May be that model is no longer available since I did this video. I will recommend checking on the forums at powerusers.microsoft.com/ in case someone has done something similar.
if I chose to download the whisper product to my local computer, how can I use Power Automate power / power apps to leverage this or is it more advantageous in the manner you’ve described in this video?
Hi Reza, i hope you are keeping well. I have a query and will be greatful it you could help me out with it - i have large data set thru a SP list and i want to connect it to combo box as a data source inside a PowerApp input form . The problem is it won't search/pull beyond 2000 rows - is there a way to get this done ..?, if you know how - a short video of the same will be very helpful
Hi Reza, great video as usual. Is there possible to do it from text to speech? Also, still now, in AI builder in my tenant, i don't have the action card " Azure OpenAI Service". Can you tell me why? and how can i get it? Thank you
AI Builder Azure Open AI is in gated preview - learn.microsoft.com/en-us/ai-builder/azure-openai-model-papp Text to speech I have not tried hence not sure
@@RezaDorrani thank Reza for answering. By the way, with whisper, is it possible to speak in African lenguages? Or Indian languages? Or does it have lenguage limitation?
Hi @RezaDorrani would you know how to do speech to text live via Power App . I am able to do it on loca web app but struggling with azure web app service.
I dont see the New Azure model you showcase here. Is it coming soon or, is there a premium option I'm missing. I have all the other models. please advise. thanks mate...
Azure OpenAI Service model overview (preview) - contains information to sign up for the preview. learn.microsoft.com/en-us/ai-builder/prebuilt-azure-openai
@@RezaDorrani I am following exactly what you did in the demo, thansk for the reply. Solution found, Firefox does strange audio format, chanegd to Edge problem solved
I have a problem, I already did all the steps but when I try the application it gives me this message(Error de Whisper.Speech2Text:{"statusCode":404, "message":"Resource not founf"})
Hello Sir, How to connect container grid view gallery(Gallery available inside container)with Editform? Since i have learnt lots of things from your videos and regular watch one by one videos.. Thanks !!
Hello Sir, I dont think I have a specific video on that. I recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com/
Reza, a normal account can't access this service? Just want to make sure that the reason is not working is because I have to subscribe to a billing plan or something. Thanks
In Powerapps, I am collecting the information with pictures. I want to patch those collection to the Sharepoint's multiline column. I tried many way. All the information getting patched with table format except images. Can you guide me
I do not have a video reference on this scenario and would have to try it out to provide guidance. I would recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com
Either your custom connector is not configured correctly or your flow is expecting a lot more params. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com as I have not come across this issue
@@RezaDorrani issue resolved. Thanks for promptly response on my query. But, now I'm unable to see the option Create text with GPT on Azure OpenAI service in power automate. Is this feature for only Premium licenses?
Hi Reza i have a quick query can you just update me that Can We Run python script in power automate to get the sentiment response on the any survey that it's good or bad.
I have not done any work with python and power automate. I recommend checking on forums in case someone has done something similar powerusers.microsoft.com
@@RezaDorrani I understand that. I actually built the connector step-by-step alongside your video. However, when I tested the app it returned an error. I have not paid the $20 per month for Chatgpt plus…what I am asking is if I need to do this, in order to get the whisper, API to work
@@kevinjoseph9023 I believe chatgpt has some free monthly credits. Beyond that you would need to buy additional credits. You would need to check chatgpt documentation for the details.
I took another shot at building the connector and followed your instructions once more. Success! I have also signed up to have access to the preview features in the most recent AI model. Thanks again, mate!
Hi Raze, I am sure to understand Swagger Edit, video at 2:28. Edit " Parameters in form Data(Audio File)" Please can you explain in more details Repeat four times, name: in: requires: type: How do you define that, Please explain? Thanks Toan
Hi Reza, I tried with my Azure subscription but I can't convert speech to text, I get the error "Resource not found" Do you know what url I could use to communicate like in your example, or what API doc I should look at? Thank you very much for the samples.
It seems that RUclips won't let me copy the url of my endpoint, but I really don't know what url to use, or what is the api doc that I should use, because the official one doesn't lead me anywhere
Details are as shared in video. I have not come across the mentioned issue and hence not sure what the fix for your issue is. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com
The feature has not been removed. Rather, it’s a preview feature that currently one needs to sign up for learn.microsoft.com/en-us/ai-builder/azure-openai-model-pauto
I dont know why, but am getting this error: {error: 1 validation error for Request body -> file field required (type=value_error.missing), type:"invalid request error"}. Not sure what am doing wrong? Does any1 know? I tried to recreate the steps numerous times
Power apps has a pen input control. You can save image from that. Then use AI builder OCR to get the text. I do not have a video reference on this though.
I have not come across the mentioned issue and hence not sure what the fix for your issue is. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com
Hello Reza, I am getting error as Invalid File Format when I click on button to generate the text. I have tried all the supported formats it has mentioned like mp3,mp4,webm,wav,mpga,mpeg,m4a but none of the extensions are working. I am trying from Windows PC and using MS edge. Please help
Video description has link to swagger file. You can compare that with yours. I have not come across the issue you are facing hence Im not sure. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com
Hello mr. Reza I am looking for way to connect two smart phons without using net for example, Farmer who work at farm need to prove the time attend, so if the boss has powerapp application on his phone which receive signal from worker’s phone to prove attendant time So can you help me to make this app Thanks
I do not have any video references on this and a scenario I have not come across. I will recommend checking on the forums at powerusers.microsoft.com/ in case someone has done something similar.
@@RezaDorrani Ok, now , I am looking for trigger concerns to Location like” When I enter or exit an area” which was availabel but now is not on the triggers list at power automate. So please help..
@@blutkraehe_official Power Apps will require a premium license, AI Builder will require add on license and for OpenAI you would need to check their documentation for costs.
you are good in powerapps but you are not a good teacher. you didnt even mention where did you get all of those things that you pasted. accept this as a constructive criticism for your future videos
Please note: Microphone control output format for mobile is different and depends upon device used. For Canvas App Microphone control, the audio formats are 3gp for Android, AAC for iOS, webm for web browsers. You will need to check the format before sending it to Whisper and convert if needed (e.g. coming from mobile device?) You could use CloudConvert connector to convert audio file type - learn.microsoft.com/en-us/connectors/cloudconvert/
Hahaha, the only reason I wanted to use Whisper over Azure was voice supporting format😮. I am already using azure and cloudconvert. 😂 Thanks for your note Reza!
Awesome video! I followed very close, but the response from the OnSelect results in a blank .text variable. Looking at PowerApps Monitor, the request "body" shows `filename=\"file_5dcd1b8d-827a-4c8b-bb45-57d58250e96b.webm`. Note the file extension. I'm using Chrome, also tried in Edge, still .webm. Firefox saves audio in .ogg, so that doesn't work either. Which browser are you using?
@@seadude Im using edge
@@RezaDorrani thanks! Your guide is so much easier than the other guy Ashiq...but how do you upload an audio recording and then convert it into a text document please ? that would be so useful !
@@dantelum I am not aware of what others are doing. its a simple attachment control in power apps. I have done several videos on attachments control.
Even before watching this video I need to thank you for taking time creating this video back to my request 4 weeks ago. And I should admit that it is more than I was expecting. . Much appreciated!
Thanks for the idea!
Allah bless you...
From Last 3 months I am trying to learn SharePoint, power apps and Power Automate.
Sir, Your videos are most helpful.
Thanks and welcome
I didn't think this was possible (despite trying all last weekend) due to the problems with sending multipart forms in a custom connector. This is such a powerful upgrade to my understanding Reza, thankyou. The GPT use cases are huge, and just in time for my talk to the Scottish Power Platform Usergroup tomorrow.
Thanks again Reza, your content is always excellent.
Most welcome Tim!
The possibilities are endless.
Yes Tim, the same to me!
wow, great presentation, you have covered so much in so less time and very prescriptive
Glad you liked it!
Simple and on to the point. Awesome content Reza.
Appreciate it!
As usually this video and your explanation is awesome. Thank you very much for sharing such a nice use case related post. No words to praise your explanation...
You are most welcome
I am so happy, it works, never thought i can get this to work..... thankyou!!!!!!!!!!!!!!!!🤣
Glad it helped
Thank you for such contents. I learned alot from your videos. Created a project also, but recently got know about multiplexing.
Could you create a video explaining it.😅
Microsoft documentation on licensing has the info on multiplexing. I have no plans to make a licensing related video.
Nicely explained Reza ❤
Thanks
Hi,
I am a beginner in PowerApps. I am trying the Whisper (Speech to Text) connector by following your video. Please may you tell where to get the Bearer API while connecting the connector in PowerApps.
Thanks
Video description has link to my AI playlist. It has videos that shows how to get the token.
Many thanks , Reza
Most welcome
Mind blowing🎉
Thanks
Awesome video Reza
Thanks
Hi Reza, Thanks for the tutorial. Now I want Text to Speech can you do it for us?
Thanks!
I have no plans for text to speech unless I get a lot of requests on this topic
Reza, this is amazing, thank you yet again! Would you be able to explain how to create a Power Automate flow that sends an audio file uploaded to SharePoint to Whisper for transcription?
Welcome!
I do not have a video reference on it and something that’s not in my planned list of videos.
I make videos when I receive multiple requests on a topic. Every 6 months I ask subscribers for topic suggestions on Community Tab of my channel. Most voted idea gets added to my backlog. Make sure to post this idea whenever I post the next topic suggestion post.
Thanks Reza - one thing you didn't mention was where/how to get the API key. Please can you offer some guidance on this, much appreciated
I have covered that in other videos related to GPT and Power Platform hence did not want to repeat the steps. Check my playlist on GPT and Power Platform.
Great stuff, Thank you! quick question what is the size limit in minutes ? Thanks Again
I have not calculated the size limit in minutes :)
Hi thank you again for this video. You didn't explain how you did o the process to convert "attach audio file" to text and the process to convert audio to image. Can you share it with us o did you already make video on those two process?
The attach audio file option uses attachments control. You can simply grab the value from that control and provide it to the API call. I have done videos on attachments control. You can see how to get value from there.
Converting from text to document after that is same as how it is explained in video.
Hey Reza
After clicking the speech to text button I got error message as "The model 'Whisper-1' doesn't exist"
Depends on how you have created the custom connector. I will have to look at your App in action to provide guidance. I recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com
Thank you very much, Sir. I'm hoping you'll teach us how to use this API to create, search, filter, and email items in SharePoint lists and Dataverse. This is to ensure that our Apps are also accessible to our impaired and blind clients.
Thanks.
I make videos when I receive multiple requests on a topic.
On Community Tab of my channel, Im currently running a topic suggestion option.
Folks suggestion ideas and most voted idea gets added to my backlog. You can try posting this idea there.
Reza 🎉 this will be so useful for our clinicians. Thanks so much!!! I only have one question once I create the connection to the api can I used it as a component in a web portal using other language or it has to be used only in the power platform?
Most welcome!
Video is related to Power Platform. You can use it in Power Apps and flows or call it from bots. Power Apps can be added to Microsoft Search, SharePoint, Teams etc.
Thanks Reza! I don’t have access to the Azure OpenAI Service
@@marienaalvarez8235 Its in gated preview. Its coming soon. You could use the ChatGPT APIs or completions API. I have done videos on those as well.
Bester Mann!
Thanks
Hi Reza, at 3:29 when I add my custom connector, I am not prompted for an API Key. Hence, when I run the app, I encounter 'invalid_api_key error. How to resolve this? I'm using Edge browser.
I have not come across the mentioned issue and hence not sure what the fix for your issue is. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com
Such a great video. I did get this to work on my desktop but not on my iPhone. It says "Invalid file format" any suggestions?
I have not come across this issue & hence not sure what the resolution is. I recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com
Hi Reza, i can't find attachment control if not in forms so i didn't quite understand how to convert audio to text from attachment, it would be very useful! Thanks!
1 - attachments control - shown in this video ruclips.net/video/QUwzvUXgj58/видео.html
2 - How to convert - is as shown in current video
Thanks Reza. Video is very helpful. I am getting error on content type. Where did can I get the content type? Thank you so much for your help in advance.
I have not come across this issue and hence not sure what is the cause for your issue. I recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com
This is amazing!! What are some use cases?
Use case is in video. I have more videos planned which will showcase other scenarios.
Hi Reza, its awesome video. I converted speech into text using whisper. Thanks for that. But I couldn't find AI Builder action 'create text with GPT on Azure Open AI Service'. As it is preview, its not available for my tenant or any configurations required?
AI builder is in gated preview.
Many thanks , Reza
Really appreciate all your efforts!
I have a question from cost prospective I have E5 license is it enough for ai models or there’s extra cost also for openAi API do you have any budgetary figures to use it within organization?
Many thanks again and again
learn.microsoft.com/en-us/ai-builder/administer-licensing
You will need a premium license for power apps + you would also need to take into account OpenAI costs related to API calls to their service.
Hi Reza, This is very helpful, thanks for posting. I don't see AI Builder for Create text using GPT. Seems it is only available as a preview feature in certain locations. Is it true? If so, any idea which region it is available and if there is any process involved in getting access to it (other than the Azure OpenAI, I already have subscription and Azure OpenAI Access). Thanks.
Check AI Builder documentation. I believe its available in US and certain other select regions only.
Awesome Video, Reza. Right what I was looking for 🙏🏻
One question has occured: As I am working in MacOS/IOS there seems to be a problem with the audio format while recording speech to text.
Is there any solution with wbmp-audio-format in MacOS environments? Thx for your help.
Just saw your pinned comment on this - will check out
Check pinned comment :)
Another question, is there any way to specify what kind of model (tiny, medium, large) and language to use? Thanks
Probably possible, but not something I have explored
Hi, please help me on this, when you used that connection in power app which API key you used there? (Bearer
I think I showed it in these videos
ruclips.net/video/F3OpKuy8wB8/видео.htmlsi=hc4efM-P5BgQ4AlC
ruclips.net/video/mzZLkf8pz9U/видео.htmlsi=kcFaYjYiWbJdSQnr
Great, Training video !!! do you have video for translation using chatGPT and Powerapps/teams?
Thanks!
I do not have a specific video on that scenario.
@@RezaDorrani Thank you for your reply... any plan to make? Please advise, Thank you
@@priyank1818 I make videos when I receive multiple requests on a topic. Every quarter I ask subscribers for topic suggestions on Community Tab of my channel. Most voted idea gets added to my backlog. Make sure to post this idea whenever I post the next topic suggestion post.
@@RezaDorrani Sure will do, Thank you
Hi Reza, Thanks for the video, But Create text, answer the question (Azure OpenAI service) is not available in my tenant, Is it not released publicly?
Its in gated preview.
This was super helpful! Can you do a new version that links into Azure Speech Services instead?
Thanks!
I make videos when I receive multiple requests on a topic. Every 6 months I ask subscribers for topic suggestions on Community Tab of my channel. Most voted idea gets added to my backlog. Make sure to post this idea whenever I post the next topic suggestion post.
@@RezaDorrani Ok, but to use Whisper via Azure is the much of a change in the approach?
@@wrayr Not sure as I have not tried it. Api would be different.
@@RezaDorrani I'm 95% there. Just hitting problems with the URI right now :(
@@wrayrKeep trying.
Il also recommend posting your issue on forums
powerusers.microsoft.com
Thanks for the video ! I have error : SpeechToTextfailed: _ScrubbedsensitiveData_
I have not come across this issue hence not sure what the fix is
Hello. good video, I have a question from where i get my API key that i need put in API Key
I got that from chatgpt by signing up for premium subscription.
Thanks Reza, super cool Video! I never thought working with OpenAI was that easy.
For longer audios, I am very interested in your example with the audio upload using the attachment control. I am having trouble in building a JSON of my uploaded audio.
Do you have a video specifically on that example?
I found a solution, save the attachment to an image control using First(DataCardValue2_1.Attachments).Value.
Then use the image control in the JSON function as Set(tempJSON, JSON(Image2.Image,JSONFormat.IncludeBinaryData))
If anyone else had the same issue. Anyway great content Reza :)
Most welcome!
I do not have a video on that specific example.
@@raffaelfrommenwiler6312 Hi, do You remember how? I'm having the same problem but I didn't understand the last part of your comment. Thanks a lot!
@@paolodeluise sorry for my late answer, i was on vacation ;)
I dont know if this is the most elegant solution, but i used an image control "img_Rec_Uploaded" to save the content of the uploaded audio.
1. At the image control add this: Image = First(att_Rec_Uploaded.Attachments).Value
2. At the attachment control add this: OnAddFile = Set(tempJSONUploaded; JSON(img_Rec_Uploaded.Image;JSONFormat.IncludeBinaryData));
Set(StringB64VarUploaded; Left(Last(Split(tempJSONUploaded;",")).Value;Len(Last(Split(tempJSONUploaded;",")).Value)-1))
3. Upload the StringB64VarUploaded variable to PowerAutomate: AIToEmail.Run(StringB64VarUploaded; First(Split(Text(aud_Rec_Uploaded.Duration); ".")).Value; "Text";JSON(ComboBox1.SelectedItems); User().FullName)
I hope this helps, good luck!
@@raffaelfrommenwiler6312 Thanks a lot for Your answer!
Can you please give a complete video on Audio file to text to document with AI demo which you have mentioned in this video at 11.19???
I do not have plans for a follow up detailed video unless many folks request for it.
Every quarter I ask subscribers for topic suggestions on Community Tab of my channel. Most voted idea gets added to my backlog. Make sure to post this idea whenever I post the next topic suggestion post.
Hi Reza, thanks for this turorial. I have one question tho... One client asked me to integrate this AI component with its SQL on-premise database to answer question regarding KPIs or filter information of tables. Do you think this is possible?
Thanks!
I have done no work with sql on-prem, but it should be possible.
@@RezaDorrani Any suggestion? :)
@@benjaminlahsen6721 I will recommend checking or posting your query on forums in case someone has done something similar powerusers.microsoft.com
Hi Reza, Its an awesome demo!! I tried this and when i use it it generates the result in the Textlabel in Hindi, is there a way i can always set the labeltext language as English? Please let me know!!
You can use AI Builders text translation service
Dear Reza, Its awesome and I would like to know how can I achieve it through power virtual agent bot which is working with MS Team.
I do not have a video reference on this scenario and would have to try it out to provide guidance. I recommend checking or posting your query on forums in case someone has done something similar powerusers.microsoft.com
@@RezaDorrani Sir is that possible to add Bing AI in fallback topic without using of bot composer. under power virtual agent bot.
@@zafarshamsi-786 Sir I have done limited work on PVA.
I recommend posting issues/queries on forums at powerusers.microsoft.com
Thank you Reza. If I record the audio on an iPhone, I get an error because the file format is WBMP but the iPhone saves audio in another format. How do you solve this?
Format for mobile is different and depends upon device used. For Canvas App Microphone control, the audio formats are 3gp for Android, AAC for iOS, webm for web browsers. You will need to check the format before sending it to Whisper and convert if needed (e.g. coming from mobile device?) You could use CloudConvert connector to convert audio file type - learn.microsoft.com/en-us/connectors/cloudconvert/
@@RezaDorrani you are right, the file format of the recording on iOS is AAC and the whisper api doesn't support it. Using CloudConvert means that the audio recorded is sent to a third party in addition to OpenAI and Microsoft and that could be a privacy concern. Do you have any other idea?
@@enderhh I do not have other ideas at the moment.
Hi @RezaDorrani, following up I'm getting error, on the text label property "The model Wishpre-1 doesn't exists" can you please suggest
May be that model is no longer available since I did this video. I will recommend checking on the forums at powerusers.microsoft.com/ in case someone has done something similar.
if I chose to download the whisper product to my local computer, how can I use Power Automate power / power apps to leverage this or is it more advantageous in the manner you’ve described in this video?
I am not aware of any download version of whisper api. All I did was leverage the chatgpt api from custom connector.
Hi Reza, i hope you are keeping well. I have a query and will be greatful it you could help me out with it - i have large data set thru a SP list and i want to connect it to combo box as a data source inside a PowerApp input form . The problem is it won't search/pull beyond 2000 rows - is there a way to get this done ..?, if you know how - a short video of the same will be very helpful
ruclips.net/video/zPM0Q53L3sE/видео.html
There is a workaround if I play the app from a IOs device? I'm getting the error "Invalid file format."
I have not experienced that issue so not sure about workarounds
Hi Reza, Thanks for the wonderful video, need to know, from where to get API Key at location 3:37
I have shown that in other AI videos.
Check ruclips.net/p/PLTyFh-qDKAiFbfKQPlEWNg5OWOXbIljri
Sir. have you used free API or need to create resource ?? How did you get API key ?
How to create api key - ruclips.net/video/F3OpKuy8wB8/видео.htmlsi=fM8qGQIHat9CTYu9
Requires chatgpt plus subscription
Hi Reza, how did you achieve the file upload control to text? and can this work with Transcriptions? Thanks!
I think that’s what I showed in video.
This works with audio files.
I’m sure there would be a way with transcriptions but I have not tried it.
Hi Reza, great video as always. I have a query. I am following your video but Swagger Editor is not loading. Please may you tell a solution.
Thanks
I have no idea why it would not load.
Its due to your browser 3rd party cookies. Click the "eyeball" icon near the right side of the address bar (in Chrome)
@@seadudeThanks alot
Hi Reza, great video as usual. Is there possible to do it from text to speech? Also, still now, in AI builder in my tenant, i don't have the action card " Azure OpenAI Service". Can you tell me why? and how can i get it? Thank you
AI Builder Azure Open AI is in gated preview - learn.microsoft.com/en-us/ai-builder/azure-openai-model-papp
Text to speech I have not tried hence not sure
@@RezaDorrani thank Reza for answering. By the way, with whisper, is it possible to speak in African lenguages? Or Indian languages? Or does it have lenguage limitation?
@@huyfe1347 OpenAI documentation has the list of languages listed
@@RezaDorrani Can i have the documention link?
Hi @RezaDorrani would you know how to do speech to text live via Power App . I am able to do it on loca web app but struggling with azure web app service.
What I showcase in video is a live speech to text live via Power App
Hello reza , can you explain in detail how to convert by uploading audio file to text in powerapps
No plans for a follow up video on this topic. I have covered the basics in the video and also demoed the file upload part.
@@RezaDorrani ok thanks. Is there any reference document to do this it will be helpful
@@sitaramthanneeru5491 I am not aware of any reference documentation for the same.
We have azure AI - wisper and i cant use this for that. Do you know how to convert the config to that ? =)
I don’t know
I dont see the New Azure model you showcase here. Is it coming soon or, is there a premium option I'm missing. I have all the other models. please advise. thanks mate...
Azure OpenAI Service model overview (preview) - contains information to sign up for the preview.
learn.microsoft.com/en-us/ai-builder/prebuilt-azure-openai
im getting this error
Whisper.Speech2Text failed: {
"error": {
"message": "Invalid file format. Supported formats: ['m4a', 'mp3', 'webm', 'mp4', 'mpga', 'wav', 'mpeg']",
"type": "invalid_request_error",
"param": null,
"code": null
}
}
Error is self-explanatory. What file type are you supplying? It must be one of the supported formats.
@@RezaDorrani I am following exactly what you did in the demo, thansk for the reply. Solution found, Firefox does strange audio format, chanegd to Edge problem solved
I have a problem, I already did all the steps but when I try the application it gives me this message(Error de Whisper.Speech2Text:{"statusCode":404, "message":"Resource not founf"})
I have not come across this issue hence not sure. Video description has link to sample swagger file which you can try.
Hello Sir,
How to connect container grid view gallery(Gallery available inside container)with Editform?
Since i have learnt lots of things from your videos and regular watch one by one videos..
Thanks !!
Hello Sir,
I dont think I have a specific video on that. I recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com/
@@RezaDorrani Request you, Please help me for same
@@swastikviewofficial2168 Forums is the best place to post issues/queries
Reza, a normal account can't access this service? Just want to make sure that the reason is not working is because I have to subscribe to a billing plan or something.
Thanks
Custom connector is a premium feature.
ChatGPT API requires a premium billing plan.
Oh ok thanks for clearing that out
In Powerapps, I am collecting the information with pictures. I want to patch those collection to the Sharepoint's multiline column. I tried many way. All the information getting patched with table format except images. Can you guide me
I do not have a video reference on this scenario and would have to try it out to provide guidance. I would recommend posting your issue/query with screenshots on the forums at powerusers.microsoft.com
Getting error - On Button On select code.
Invalid Number of arguments: received 4, expected 0-1.
Either your custom connector is not configured correctly or your flow is expecting a lot more params. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com as I have not come across this issue
@@RezaDorrani issue resolved. Thanks for promptly response on my query. But, now I'm unable to see the option Create text with GPT on Azure OpenAI service in power automate. Is this feature for only Premium licenses?
@@deepaksartape9502 Its a preview feature not released on all tenants yet. It will be a premium feature whenever it goes GA.
Hi,Reza its awesome i try to do same but api key fails what is the api key?
You will need to get your own api key. I have done many videos on chatgpt. Link is in video description. It will show you how to get your own key.
Hi Reza i have a quick query can you just update me that Can We Run python script in power automate to get the sentiment response on the any survey that it's good or bad.
I have not done any work with python and power automate. I recommend checking on forums in case someone has done something similar powerusers.microsoft.com
Do I have to pay for the $20 a month version in order to leverage this API in this fashion
Any custom connector is a premium feature
@@RezaDorrani I understand that. I actually built the connector step-by-step alongside your video. However, when I tested the app it returned an error. I have not paid the $20 per month for Chatgpt plus…what I am asking is if I need to do this, in order to get the whisper, API to work
@@kevinjoseph9023 I believe chatgpt has some free monthly credits. Beyond that you would need to buy additional credits. You would need to check chatgpt documentation for the details.
I took another shot at building the connector and followed your instructions once more. Success! I have also signed up to have access to the preview features in the most recent AI model. Thanks again, mate!
Hi Raze,
I am sure to understand Swagger Edit, video at 2:28.
Edit
" Parameters in form Data(Audio File)"
Please can you explain in more details
Repeat four times,
name:
in:
requires:
type:
How do you define that,
Please explain?
Thanks
Toan
Video description has link to reference
Swagger file.
Hi Reza, I tried with my Azure subscription but I can't convert speech to text, I get the error "Resource not found"
Do you know what url I could use to communicate like in your example, or what API doc I should look at?
Thank you very much for the samples.
It seems that RUclips won't let me copy the url of my endpoint, but I really don't know what url to use, or what is the api doc that I should use, because the official one doesn't lead me anywhere
Details are as shared in video.
I have not come across the mentioned issue and hence not sure what the fix for your issue is. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com
@@RezaDorrani Thanks Reza, I'll try that
Hi Reza, is there any way to send the request to Whisper for a different language?
There is a translation api with whisper. Check the documentation of whisper at openai api
form where can I get the API key you use it at 3:35 ?
Covered in ruclips.net/video/mzZLkf8pz9U/видео.html
i think microsoft have removed this Feature i cannot see in my ai builder option for azure opean ai service model
The feature has not been removed. Rather, it’s a preview feature that currently one needs to sign up for
learn.microsoft.com/en-us/ai-builder/azure-openai-model-pauto
@@RezaDorrani thank you!!
I dont know why, but am getting this error: {error: 1 validation error for Request
body -> file
field required (type=value_error.missing), type:"invalid request error"}. Not sure what am doing wrong? Does any1 know? I tried to recreate the steps numerous times
I have not comes across this hence not sure. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com
Hi i am getting a error "you exceeded your quota please check plan"
You would need a premium plan for OpenAI API. Check OpenAI documentation for details.
@@RezaDorrani is there perhaps a workaround or a way I can just use it for testing ?
@@justindavid8401 Im not aware of any workarounds apart from the premium plan
How to perform iPad pencil writing converted into text?
Power apps has a pen input control. You can save image from that. Then use AI builder OCR to get the text. I do not have a video reference on this though.
It shows %3C%21--# scrubbedSensitiveData_error
I have not come across the mentioned issue and hence not sure what the fix for your issue is. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com
Hello Reza, I am getting error as Invalid File Format when I click on button to generate the text. I have tried all the supported formats it has mentioned like mp3,mp4,webm,wav,mpga,mpeg,m4a but none of the extensions are working. I am trying from Windows PC and using MS edge. Please help
Video description has link to swagger file. You can compare that with yours.
I have not come across the issue you are facing hence Im not sure. I recommend posting your issue with screenshots on the forums at powerusers.microsoft.com
@@RezaDorrani Swagger File is correct. The issue is with the File extension. I am stuck! Shared the screenshots in the forum
@@prasbaby No idea bro as I have not faced this issue.
@@RezaDorrani im also receiving this issue, can you test again in your app and see if it still works?
@@dominicliu2065 Works fine for me. Not sure if something has changed on API side.
anyone have any ideas on how to convert the text to a Dataverse record with the appropriate fields for the record filled out automatically?
I did a recent video on extracting text from email. You could use similar concept for output of speech to text and see whats possible.
@@RezaDorrani such a legend, thank you! amazing content
Thanks
Where do I get the API key?
Video description has link to my AI and power platform playlist. I have done videos where I have shown how to get the key.
connector free? it's possible?
Not free. Requires premium licensing.
Hello mr. Reza
I am looking for way to connect two smart phons without using net for example,
Farmer who work at farm need to prove the time attend, so if the boss has powerapp application on his phone which receive signal from worker’s phone to prove attendant time
So can you help me to make this app
Thanks
I do not have any video references on this and a scenario I have not come across.
I will recommend checking on the forums at powerusers.microsoft.com/ in case someone has done something similar.
@@RezaDorrani Ok, now ,
I am looking for trigger concerns to Location like” When I enter or exit an area” which was availabel but now is not on the triggers list at power automate. So please help..
@@qwertyytrewq8796 That feature has been deprecated
Is that free?
No
@@RezaDorrani oh, how much it Costs?
@@blutkraehe_official Power Apps will require a premium license, AI Builder will require add on license and for OpenAI you would need to check their documentation for costs.
you are good in powerapps but you are not a good teacher. you didnt even mention where did you get all of those things that you pasted. accept this as a constructive criticism for your future videos
Accepted
where can i get the api key?
Shown in this video 👉 ruclips.net/video/F3OpKuy8wB8/видео.html
@@RezaDorrani Thanks @RezaDorrani and thank you for sharing your knowledge!
@@RezaDorrani Is it necessary to have a paid version of OpenAI to be able to use the API key for this functionality?
@@crisantolima1423 Yes. I dont think they have a trial option.