Rime secures $24M Series A to develop voice AI models for enterprise customer calls
The voice AI startup aims to stand out with its specially trained models for customer service.
1A new player in voice AI market
Rime is a startup developing specialized voice AI models to handle customer calls. Founded in 2022 by Stanford alumni and former Amazon engineers, the company differentiates itself by collecting training data from its own recording studio in San Francisco rather than scraping the web.
2Customized training approach
Instead of web scraping, Rime gathers its own conversational data through an in-house recording studio. This enables accurate pronunciation of brand names and industry-specific terms. Its phoneme-based architecture adapts to various pronunciations, eliminating the need for customers to retrain models.
3$24 million funding round
Rime has closed a $24 million Series A round led by M13 Ventures, with participation from Twilio Ventures and Unusual Ventures. The startup previously raised $5.5 million in seed funding in May 2023. New funds will expand engineering and model development teams.
4Diverse industry clientele
Rime serves clients in food service, healthcare, airlines, and fintech sectors. Working with enterprises like Mayo Clinic and Asurion, the company claims its models increase call duration, helping secure more enterprise contracts. Its training data focuses on industry-specific terminology.
5Technical challenges and goals
Founder Lily Clifford acknowledges current voice AI still can't match traditional IVR effectiveness. Rime is shifting focus to speech-to-speech models to reduce latency and address background noise issues. This new approach will also decrease dependency on model orchestration.
6Competitive differentiation
Competing against players like ElevenLabs and Deepgram, Rime stands out with low-latency, high-reliability models. M13's Morgan Blumberg highlights Rime's technical approach in regulated environments as industry-leading. The company recently added talent from NVIDIA and Meta to its team.