hi there are you a business owner trying to decide which voice a system suits for your needs then you have come to the right place in this video I will demonstrate you two important voice AI system that is VAP Ai and Retail AI by the end of this video you will have a clear idea of which one fits for your needs saving your time and stress we will cover everything from E of use cost Integrity analytics user interface and so on so let's dive into to the video Let's Start by comparing the user interface of both vapy Ai and R AI in order to start vapi you can press dashboard then you will came to this page on the left side you can see overview platform voice Library logs when you press platform you can see assistant phone number files tools blocks Squad so I'm going to start by creating an assistant for that you can press create assistant then a popup will comes here you can change your assistant name I'm going to put it as new assistant and to kick start your assistant creation you have four templates that are appointment setup customer support inpound question answer game and PC if you wish to start with a blank template then you can go here and you can press blank template and you can create an assistant here you can see that a new assistant has been created you can see the assistant ID here when you scroll down you can see model where this section allow you to configure the model for the system and inside model you can see the first message and the system prompt first message is a message that your assistant will say first and here you can see system prompt system prompt can be used to configure the context role personality instruction and so on for your assistant now let's start R AI this is the first page of R AI you can start by pressing login so once once you have logged in you will come to this landing page here on the left side you can see agents knowledge Base phone number batch call call history billing API keys and webook and here you can see some of the invoice details and the concurrency used and I'm going to start by creating a new agent for that you can press create an agent uh I'm going to select single prompt agent here you can see that there are pre-i templates like healthare checkin notification I'm going to start from blank new agent has been created here you can see the system prompt and on the welcome message you can see a wide variety uh you can select user initiates when a remains silent until user speak first or you can select a initiates a begins with a dynamic begin message or third one that is a initiates a begins with your defined begin message so you have wide variety of Welcome message in a single prompt agent but in the case of Vy you can only pass a defined message in first message in R AI if you wish to give a defined be message you can press this and you can pass this here you can see that hello how can I help you today you can customize this according to your needs now when we check uh you can see that all the function knowledge base speech settings call settings Etc are itself in the single uh page itself but in the case of vapy you can see that all the things like model transcriber everything is in different different set ction now let's talk about the provider here you can see that open a provider is there along with the anthropic models but in the case of vippy you can see that there is a wide variety of providers are there and another important thing is the custom llm when we go to create an agent here you can see that create an agent and inside this you can see custom llm if you want a custom LM rather than open AI or anthropic then you can select custom llm and here you can enter your custom llm URL so the same thing is itself in vapy also in vapy section when we come to provider you can scroll down and you can see custom llm and here you can pass the custom llm URL now let's check for multi- prompt agent multi- prompt agents are used for lengthy calls and complex task here you can see that there are templates like d distry schedules lead qualification or you can start from blank here you can see that the important thing is the prompt tree in the multi- prompt agent uh in Rell AI the agent Builders are the single prompt and the multi- prompt but in the case of app AI the agent Builders are single prompt and block agents here you can see that when I press edit promp tree there will be different stage so it is just a workflow in every stage you can mention the prompt to be worked on that particular stage along with tools and custom function so I will show you an example see this is a prompt here you can add tools and here you can add the custom function also in the case of wapy you can see that uh when you SC when you see left you can see blocks in order to create new blocks workflow you can press this and you can give give any the name and you can select the assistant now you can press create workflow you can see that a new block has been created where you can custom the instruction and also you can customize the expected output data and you can add this another step by pressing add step and here you can uh choose conversation or you can choose tool call or you can use workflow or you can use an existing block itself so that is the difference between uh rell's workflow and V's workflow this is just an overview if you wish to create a VAP a assistant or a assistant we have made videos about VY and return you can can check out our YouTube channel for more details let's discuss about speed of fappy A and R AI here you can see that I have created same agent in R Ai and VY itself here in R AI I'm selecting the model as GPT 40 mini I'm seling the same model in B itself that is gp4 mini now let's uh test a web call hello this is Mary from Mary's Dental how can I assist you today uh hi Mary my name is John hey John nice to meet you so what's the reason for your visit yeah I need to have an appointment got it when do you want to come in any specific date and time in mind yeah I need to have a schedule on tomorrow 2 p.m. all right so that's tomorrow at 2 p.m just a double check can I get your full name and the purpose of your appointment yeah my name is John and my purpose is uh teeth pain ouch teeth pain is no fun so just to confirm we've got you John for a teeth pain appointment tomorrow at 2 p.m. sound good yes awesome you're all set John we'll see you tomorrow at 2 p.m. don't forget to bring your smile just in case anything else I can help you with no thank you no problem John thanks for calling have a great day and see you tomorrow now let's test FY hello this is Mary from Dental how can I assist you today hi my name is John hey John what can I do for you today yeah please book an appointment sure thing John what are you coming in for uh I have severe teeth pain ouch that doesn't sound like fun when would you like to come in uh I would like to come in for tomorrow 2 p.m. got it John you're all set for tomorrow at 2: p.m. we'll see you then thank you you're welcome see you tomorrow according to the testing retel a has a latency of 1.60 seconds average but VAP a have a latency of 1.93 seconds average now let's talk about the voice quality of both the agents when we come to R AI you can see that the main voices are from play HD 11 labs and open AI when we come to W you can see that there are wide varieties of voice providers starting from kesia to Tas now let's test with a web called the quality of both the agents hello this is Mary from Mary's Dental how can I assist you today hi my name is John hey there John nice to meet you what's the reason for your visit today now let's ask for Wy both these things are in 11 Labs only hello this is Mary from Mary's Dental how can I assist you today hi my name is John nice to meet you John what can I do for you today yeah from the testing which I found that R AI have more voice quality as compared to W AI where both these are from same provider that is 11 Labs but you can increase the uh quality of Sound by additional configuration in bapy you can set up the background sounds as off office default and when you scoll down you can uh increase the stability you can customize the clarity plus similarities style exageration optimize streaming latency and use speaker boost Etc here when we come to the retail AI you can uh customize the voice speed voice temperature voice volume and when we come to speech settings you can select the background sound from coffee shops conventional Hall summer Outdoors Mountain outdoor static noise call center none in thaty there is only three main options for background sounds but R AI have wide variety of background sounds and you can also customize responsiveness Interruption sensitivity and you can enable back channeling boost keywords enable speech notification Etc now let's compare the cost distribution of vapy AI and Ral AI in the case of vapy AI here you can see different models are there the cost is depends upon the model we used here you can see that the model which I have selected is gp40 mini and here you can see that about1 0.11 per minute is the cost this is the cost distribution VAP itself have a fixed cost that is USD 0.05 where rle a doesn't have a fixed cost on the on the cost distribution you can see that it has a fixed cost for WAP itself along with deep gr and the GPT model and 11 laps so total it has about $0.11 per minute when we select another model you can see that the cost changes that is about1 0.2 per minute now let's check for retail AI in the case ret in the case of ret a you can see that the cost per minutes is about do 0.76 per minute in dollars here I have selected the model as GPD 40 mini now I will select another model here you can see that the cost has been changed into 0.2 per minute dollar here you can see the cost per minute distribution so the pricing of retel a basically for voice AI is1 0.0 sub per minute and for vippy it is about $1 0.066 plus do 0.05 per minutes that is fixed platform fee prices varies if prompt is long and for phone number rle a have a fixed price of12 per month and for vapy also it have a fixed price of $2 per month thus we can conclude that retel is more cost effective especially for complex agents and for higher call volumes now let's compare R Ai and VAP AI in terms of Integrations and functions both platforms integrates with external services but here in case of R AI you can see that there's a wide varieties of functions here like n call call transfer check calendar availability hook on calendar press digit custom function Etc but in case of vapy you can see that VY Rees manual setup for similar functionalities via third party tools like make sapure Etc so this is considered to be as a drawback of papy but in the case of retel AI you can see that retel has an build-in cal.com integration for scheduling making it more convenient for customer facing roles here you can see that you can check the calendar availability on cal.com by passing the APA key of your cal.com account along with the event type ID and here also you can select the time zone after passing this you can select book on calendar here also you can pass the API key of your cal.com and you can pass the event type ID and the time zone so with the help of R A you can easily schedule an appointment on cal.com I have created a detailed video about creating custom functions and tools on Wy Ai and R AI in our previous YouTube videos for more details you can refer those videos now let's compare wapy Ai and R AI in terms of testing and Analysis vapy offers detailed analysis and a success evaluation tool for assessing conversation quality here you can see that there is summary is there success evaluation is there and structure data is there summary is uh used to summarize the call you can pass the prompt here and success evaluation will evaluate whether your call was successful you can pass the prompt user to evaluate the call success call is whether successful or not and when you scroll down you can see that there are success evaluation rubrics this will enforces the rubrics of the EV valuation upon the success evaluation here there are numeric scale descriptive scale checklist metrics percentage scale Leed scale and automatic rubric when you scroll down you can see structure data this will extract the structur data from call conversation you can pass the prompt here now let's check for retel AI here in the post call analysis here you can see that there is post called Data retrieval which will Define the information that you need to extract from call here you have text selector Boolean numer retel provides a simpler set of analysis but its main feature is that test llm here you can see test llm where test llm feature allow you to simulate voice calls in chat board mode streamlining prompt testing now let's discuss about multilingual language in the case of R AI it only supports 90 language but in vapy AI it support about 35 Plus language so VY have more weightage on multilingual language compared to Ral AI here you can see that these are the language supported by ritual AI but when we come to wppi when we come to transcribe language you can see that about 35 Plus languages are supported now let's compare in terms of fastest multilingual agent as compared to W little a is the fastest multili agent now let's test this here you can see that I have already created an agent known as M and here I'm going to select the model as GPT 40 mini real time now let's test this hello this is Mary from Mary's Dental how can I assist you today hi my name is John can you please speak in Hindi hello appoint problem yeah routine checkup great yes R AI is the fastest multilingual agent as compared to any voice a system we have explored the key differences between R Ai and B AI in terms of user interface speed voice quality sound Integrations testing analysis Etc if you're looking for in-depth analysis a variety of voices and more customization then B AI will be the best choice for you on other hand if you need a more cost effective solution with build-in scheduling and faster testing then rle could be the best choice for you ultimately both the platform has their own unique strength so it is important to choose the right one that aligns with your needs thanks for watching now you have a better understanding on both the voice A system that is VY Ai and ritual AI including how they perform if you're planning to create a voice a system for your company then mq automation is there for you whether you care more about flexibility speed or keeping your course down then there is a solution here for you drop a comment to let us know which platform you are thinking or if you have any questions don't forget to check out our previous videos regarding voice AA systems if you found this video helpful then please give us a thumbs up and subscribe for more exciting videos see you next time thank you