NovaSynth by Noveum
Test your voice agent on the callers you can’t stage.
455 followers
Test your voice agent on the callers you can’t stage.
455 followers
Simulate realistic callers at scale with custom personas, scenarios, interruptions, noise, accents, and network conditions. NovaSynth runs those calls against your voice agent, scores audio and transcripts across 30+ dimensions, and surfaces the failures and fixes that matter to your team.
Free Options
Launch Team / Built With
Render: Get up to $100K in startup creditsThe fastest way to launch and scale your startup in the cloud
Promoted
Hi Product Hunt 👋
I’m Additi, co-founder of Noveum AI.
Voice agents have guardrails, scripted tests, and observability. The caller is still a variable
Real callers don’t work that way.
- The Interrupter: Cuts you off mid-sentence, talks over the agent, then demands an immediate answer.
- The Second-Guesser: Agrees, then changes their mind twice and asks for a completely different option.
- The Hard-to-Understand Caller: Speaks softly, switches pace, uses an unfamiliar accent, and repeats themselves.
- The Distracted Ones: Calling from a crowded, noisy place with people talking, traffic, and constant interruptions around them.
- The Frustrated Caller: Starts impatient, gets increasingly annoyed with every extra question, and is close to hanging up.
- The Unstable Connection: Audio drops, words get clipped, pauses appear unexpectedly, and the caller repeats themselves.
And you can’t keep relying on a human tester to recreate every one of those situations.
Why NovaSynth Exists
A year ago, Shashank and I were working on voice agent evaluation when someone said:
“The happy paths are working fine. I pray the agent works well when a weird customer picks the call."
That question stayed with us. It became the starting point for NovaSynth, a pre-production testing platform built around simulated callers.
KEY FEATURES
→ Build the caller, not just the test: Create personas and scenarios around accent, mood, behavior, interruptions, noise, network conditions, and caller intent.
→ Mix and match personas and scenarios: Create different callers for different situations, then test any persona against any scenario as your agent evolves.
→ Run real voice interactions: Simulate calls through real audio and telephony paths, capturing voice-specific issues such as latency, interruptions, and audio quality.
→ Evaluate beyond the transcript: Score calls across 30+ audio and 100+ transcript scorers, plus the business KPIs your team cares about.
→ Validate fixes before shipping: Use NovaPilot to recommend fixes, then backtest and re-run them against your own calls and traces.
We’re here today, Shashank and I will be around all day to answer any of your questions or doubts.
Start a Free Trial: https://bit.ly/4zOcEXT
Book a Demo: https://bit.ly/4gtBXae
And I’m curious:
What’s one caller you wish you could test your voice agent against, but have never been able to recreate?
@additi Many congratulations Harkirat and team on the launch! :)
How I met the makers?
I first worked with @harkirat_singh3777 at previous companies he was part of, and when he joined Noveum he reached out and asked me to come check this out too.
What is the product?
NovaSynth is a testing platform for voice agents that simulates realistic callers with different personas, interruptions, accents, noise, and network conditions, then evaluates the results across audio and transcript dimensions.
Why I endorse it?
I endorse it because it solves a very real problem for teams building voice agents: the gap between happy-path testing and messy real-world calls. It feels practical, thoughtful, and genuinely useful for catching issues before production.
Give it a spin!
@rohanrecommends Thank you so much for checking out NovaSynth and writing this up. Really appreciate the support. ❤️
A big part of what we’re trying to solve at Noveum is what happens when voice agents go beyond the happy path. That’s where things get messy, and where a lot of testing tends to fall short.
Glad that came through in your experience with NovaSynth. Thanks again for backing the launch!
Super special day for us ❤️
NovaSynth is one of those products we’ve genuinely built with our heart. I’ve been lucky enough to be one of the makers behind it, and seeing it finally out in the world feels kinda unreal.
We’ve spent a lot of time thinking about one simple problem: voice agents work great in perfect test cases, but real callers are anything but perfect.
Interruptions, weird scenarios, bad networks, noise, accents, frustrated users, people changing their minds mid-call... NovaSynth is our attempt to make all of that testable before your agent actually hits production.
There’s still so much we want to build, but today I’m just really happy to see this ship.
Would genuinely love to hear what you all think, especially the weirdest caller scenario you’d want to throw at a voice agent :)
@_pulkitxm Thanks for your contributions. Your job today is to make sure the platform can handle the traffic.
Let's go!!!
@itsshashank yep definitely 🙌
Anything that makes AI evals easier is a big blank space across modalities. This is a fantastic space to be building in - kudos to you folks! Are you thinking of sticking to voice agents, or would you be doing persona testing on chat based apps at some point too?
@prajwal_prakash thanks. We we are already doing chatbot evaluation and also have an HTTP endpoint that lets you do this for chat apps too. If you’d like to try it, happy to jump on a quick call and help you get set up.
Congrats on this launch builders , what's one distinguish thing about noveum that you are proud of building ?!
@easeops Well, in NovaSynth, we don't just tell you the error. We actually tell the customer how to fix it. We tell them the exact backtested system prompt changes, model swaps, everything.
Interesting, this will be helpful in testing the new voice agents
@umang_gupta9 Thank you. Do try it out.
A great way to ensure in prod our voice agents give comparable quality as we got in our pilots !!!
Pretty great usecase actually. we've also seen these requirements and need for a specific tool among our clientele.
@pr_khar do get in touch to discuss the use cases. We run evaluation for pre-prod as well as voice agents running in production.