How Siro reduced customer complaints and support tickets by 90% with AssemblyAI

AI-powered coaching platform for field sales, Siro, services customers in industries that send sales representatives on-site to sell or perform a service, such as HVAC and home improvement. This traditional selling approach can be a big problem, Siro co-founder and CEO Jake Cronin explains, because many of the "salespeople aren't getting proper coaching."

In-person coaching simply requires too much time and capital for companies, especially with sales teams dispersed over wide regions—but skipping coaching can lead to less company oversight over sales practices and reduced revenue for sales representatives.

Cronin and his team at Siro set out to solve this problem by building a platform that would use technology to close the feedback loop and optimize in-field sales practices remotely and at scale.

{{play-video-button}}

Since the sales agents work directly in the field, Cronin knew the success of the Siro platform would depend on two main factors:

• Highly accurate speech transcription from recordings on mobile devices (e.g., cell phones or tablets)
• Highly accurate speaker diarization, or speaker identification and labeling, to automatically separate sales agents and customers on these recordings

Background noise, accents, and other factors also affect the quality of the recording, which directly impacts the quality of the transcription.

Cronin's team initially added a legacy speech-to-text provider that they hoped would fill this need. But when support tickets started building and customers began churning, it quickly became apparent that they needed a new option—which led them to AssemblyAI.

‍

AssemblyAI's Universal speech recognition model offers best-in-class transcription and speaker diarization, even in noisy, less-than-ideal recording environments, as well as low latency at 23 seconds on a 30-minute audio file. AssemblyAI's latest speaker diarization model is also 13% more accurate than its predecessor and demonstrates an 85.4% reduction in speaker count errors.

With AssemblyAI, Siro was able to reduce customer complaints and support tickets by 90%.

‍