Hajra Naz
The idea that voice will be the next big interface has picked up strong momentum. Investors are pouring billions of dollars into voice AI startups. These companies work in many areas. Some build the models. Others serve enterprise customer service. Some take meeting notes. Others focus on AI-powered dictation.
Voice AI’s “ChatGPT Moment” Hasn’t Arrived
A new model or tool arrives almost every week. Each one claims to sound human and hold a conversation like one. Reality may be different.
Shawn Wen is the CTO of enterprise voice AI platform PolyAI. He thinks voice AI hasn’t had its “ChatGPT moment” yet. That’s true even with the release of full-duplex models. These models can speak while they listen to you.
“We have reached the milestone of developing full-duplex models. The next challenge is to make reasoning very fast so that the models can fetch answers quickly and the conversation feels natural,” he said on stage at the HumanX conference last month.
Why Speed Matters
Pauses break the feel of a conversation. A person expects a quick reply. Even a short delay can make an AI agent feel stiff. Fast reasoning lets the model look up an answer and respond in rhythm. That’s what makes a call feel natural.
Read More: ChatGPT Introduces a New Interface With a More Visual Experience
Building Caller Confidence
Wen also said customer service agents shouldn’t sound robotic. They should give callers confidence that the agent can solve their problem.
“I think the next stage will be slightly different because once the voice is good enough, like, and the customer is willing to engage with them for the first two or three turns, they start to build confidence, and over time, they will feel like I probably don’t have to talk to a human if the agent can solve my problem,” he said.
The first few turns matter most. If a caller stays for those, trust can grow. Over time, the caller may stop asking for a person.
Meetings and Digital Twins
Alex Gay is the CMO of meeting notetaker Otter. He said a few steps unlock automation. The tool must identify speakers. It must capture intent. It must then write that up using the organization’s knowledge.
Otter is also working on digital twins. These could represent people in meetings. Gay said the voice must carry the same emotion as a human voice for this to work.
“If you think about the meetings that you’re in right now, the best conversations that you have are where you can have debate and strategic discussions, and when you feel like there’s a relationship that underpins it. If you aren’t able to have that with an avatar, then it’s just a q and a chatbot,” Gay noted.
Read More: OpenAI Will Start Watermarking ChatGPT Text Across the EU
Understanding and Accuracy
Voice AI models have improved. Problems remain, though. Assistants often misunderstand users. A notetaker may produce the wrong transcript or summary.
Wen thinks ASR (Automatic Speech Recognition) models often miss important keywords. That makes it hard to capture the full context.
Gay agreed. He said Otter keeps working on better transcription. He also said language is an area where voice models need to improve.
“For Otter, you know, transcription was never the end point. It was just the layer that we could start to drive some of the productivity gains on the back of. But if your original transcription didn’t have the accuracy that you needed, all the follow-up actions that you have become flawed. And the minute that starts to take action, that is wrong. You lose trust in the platform. We must continue to improve that ASR model because all of the downstream impacts are significant,” he said.
Why Errors Compound
An error in a transcript doesn’t stay small. A wrong word can lead to a wrong summary. A wrong summary can lead to a wrong task. Each step builds on the one before. That’s why accuracy at the start matters so much.
Read More: ChatGPT Will Show Visual Ads During Image Generation, Starting With a US Test
Transparency
New voice tools also raise a transparency question. Tools should tell people when they are being recorded. They should also say when someone is talking to an AI.
Otter says it wants to build trust among people in a meeting. Even when its bot isn’t present, it wants to try methods like telling everyone in the chat that the meeting is being recorded. Wen also said it’s important to make clear that people are talking to an AI on enterprise calls.
What Comes Next
Voice AI is moving fast. Speed, accuracy, and honesty look like the main tests. Companies that pass them may earn lasting trust. Those that don’t may lose it quickly.
