Automated testing and monitoring for AI voice agents.
External link for Hamming AI
San Francisco, California 94105, US
Hamming AI reposted this
Episode 3 of The Voice Loop is live. This time, I sat down with Fionn Delahunty, PM at Synthflow AI, to talk about building and scaling a no-code voice AI platform that handles millions of calls. A few things we got into: - Why no-code still wins even for enterprises - Vendor reliability - Running a remote, globally distributed team - Hiring for a Series A voice AI startup Our podcast is now on Spotify, link in the comments :)
When you're building in voice AI, you'll realize audio quality is one of the most underrated layers in the entire stack. When you analyze audio at scale and dig into thousands of real user recordings, you start seeing distortions you never planned for, not just background noise or bad mics, but strange, hyper-specific issues that only show up in real environments. Every new dataset reveals something new. Teams like ai-coustics do something counterintuitive: They intentionally destroy clean audio so models learn to survive messy, real-world conditions. And that’s exactly why, at Hamming, you can simulate the environments your voice agents will actually face, from cars and kitchens to drive-thrus and call centers, etc. If production variables are unpredictable, your testing variables have to be too.
Most people don’t realize how much raw complexity sits inside a single audio signal. In our latest episode of the Voice Loop, Sumanyu Sharma and Fabian Seipel unpack a highly technical component of audio engineering: source separation. The ability to take one audio track and break it into its individual components. Back in 2014, Fabian’s internship at Sony sparked his curiosity. Here's what he had to say 👇️
Hamming AI reposted this
The Voice Loop is back, this time delving deep into the topic of voice agent performance. Head to the comments section to see the full conversation between Sumanyu Sharma and our co-founder, Fabian.
🎙 Episode 2 of The Voice Loop is live. I sat down with Fabian Seipel, co-founder of ai-coustics, to unpack one of the most overlooked layers in the voice AI stack: audio quality. We discuss everything from simulating real-world audio distortions to tackling the complexities of human speech. We went deep on how to build an audio quality layer that actually holds up in production. Link in comments!
Hamming AI reposted this
🎙 Episode 2 of The Voice Loop is live. I sat down with Fabian Seipel, co-founder of ai-coustics, to unpack one of the most overlooked layers in the voice AI stack: audio quality. We discuss everything from simulating real-world audio distortions to tackling the complexities of human speech. We went deep on how to build an audio quality layer that actually holds up in production. Link in comments!
Hamming AI reposted this
Tool calling remains one of the biggest weaknesses of LLMs when building voice agents. A voice agent can sound natural and handle complex conversations, but if it can't reliably check inventory, book an appointment, or pull up account information when needed, the whole experience falls apart. With Hamming AI, you can test your tool calls. You can run calls where you can trigger edge cases in the API integrations to see what breaks. LLMs are getting really good at understanding intent and generating natural responses, but the execution layer (actually doing the thing the user asked for) is still where most failures happen. What's your biggest challenge(s) with tool call reliability?
In our latest podcast episode, we talked about one of the most overlooked truths in voice AI: Domain expertise beats model choice. The best-performing voice agents are shaped by the realities of the use case: the industry, the call structure, the team’s capabilities, and the regulatory constraints around the workflow. Swipe to learn more 👇️
Hamming AI reposted this
Checkout my conversation with Sumanyu Sharma about why outcomes are key 👇
We asked Lily Clifford, CEO of Rime: What really makes a voice agent successful? The answer: It’s not latency, call volume, or uptime, it’s outcomes. Listen to our latest podcast episode to learn more about measuring voice AI outcomes that actually matter.
We asked Lily Clifford, CEO of Rime: What really makes a voice agent successful? The answer: It’s not latency, call volume, or uptime, it’s outcomes. Listen to our latest podcast episode to learn more about measuring voice AI outcomes that actually matter.