Vapi Ai - Will it replace the call centres?
Conversational Voice chat bots are a rage these days and they are getting better day by day. However, will the human still have a role to play? How good are these bots?
We are hearing a lot about how jobs related to call center are at risk with emergence of voice bots. But how far are we and how good are these bots? Which companies are building them.
Today I will tell you about one such company i.e. Vapi.ai, which has been one among the platforms that are enabling companies to build and launch their own voice bots that are super friendly.
How Vapi.ai Works
Vapi.ai is designed to help businesses automate their communication processes through advanced voice AI technology. By leveraging Natural Language Processing (NLP) and machine learning, Vapi.ai can understand and respond to customer inquiries in a conversational manner. This allows businesses to handle customer support, sales calls, and other interactions more efficiently.
They provide a simple, reliable and accessible API for voice AI, with features like turbo latency optimizations, interruption handling, and scalable infrastructure. Vapi supports multiple models, voices, and backends, and offers client-side SDKs for various platforms. They are backed by Y Combinator and have been featured in several news articles. Vapi costs $0.05 per minute, with no hidden costs. They also provide a $10 free credit to get started.
Peak Into Vapi.ai Dashboard:
Gives you an estimate of the cost per call
Integrates with Twilio for quick calls
Select your own models
HIPAA Compliant and a lot of advanced features for you to try.
Advanced Voice Customisations:
Check out the prototype I have made for the company I am working, Invest4Edu . We are into making education planning easy. Helping every child become their best. Ask the voice bot more questions to try it out yourself.
Link to try: VapiAI - Demo
Flexibility and Customization
Multilingual Support: Create voice agents in a variety of languages, including English, Spanish, German, Hindi, Portuguese, and over 100 others.
Bring Your Own Models: Utilize your preferred models from providers like OpenAI, Groq, Mistral, OpenRouter, Together, and Anyscale.
Bring Your Own Voices: Integrate voices from ElevenLabs, PlayHT, LMNT, Deepgram, Cartesia, Rime, OpenAI, and Azure.
Bring Your Own Backend: Leverage Vapi's function calling capabilities to build agents with custom backend integrations, extending their functionality significantly.
Bring Your Own Surface: Use Vapi's client-side SDKs for Web, iOS, Flutter, React Native, and Python (Hardware) to integrate your voice agents into various applications.
Though it’s great!! Latency is one issue that once solved will make a huge difference here. Also, it still does feel like robotic sometimes, human emotions are still difficult, to capture by an AI properly yet.
Companies like Groq - Groq solutions are based on the Language Processing Unit (LPU), a new category of processor. Groq is the creator of the LPU and built it from the ground up to meet the unique characteristics and needs of AI. LPUs run Large Language Models (LLMs) at substantially faster speeds and, on an architectural level, up to 10x better energy efficiency compared to GPUs.
Cerebras - They claim to be world’s fastest inference. 20x faster than GPUs, 1/5 the cost, are working on bringing the latency down with advanced technologies.
Companies like Hume.ai , is making human conversation more humane. You must check it out to know the difference. Elevenlabs, OpenAI, Deepgram etc.. are not far behind as well.
So will the human agents exists or will AI take over the call centre ?
My Take: Human agents will stay but their quality of work will increase, basic queries will get addressed by these bots but when there’s a complexity, emotions are involved we still need people to solve for it.
Customer service jobs will not go away. People in these roles will get a better quality of work. Humans always love to speak to humans, especially when they are very angry, upset with service, P0 ,P1 level escalations etc…
If your customer service team is spending time answering basic FAQs those you should focussing on getting it automated. Once people start relying on these voice bots then many others follow soon.
Company: Vapi.ai
Funding: Funded by Y-Combinator around 125K to 500K USD