Indian voice AI startup Ringg has raised $10 million from Peak XV Partners as an extension to its Series A, bringing the round to $15.5 million. The company is using the funding to move beyond high-volume phone campaigns and toward more complex enterprise tasks across voice, chat, WhatsApp and browser support.
Ringg says it already processes 20 million call attempts each month. Its opportunity is shaped by a market where a Truecaller study found that more than 76% of Indian consumers prefer talking to businesses by phone, giving automated voice systems a large existing behavior to serve.
From speech models to enterprise orchestration
Ringg began as DesiVocal, a text-to-speech startup. Training proprietary speech models proved expensive, so the founders shifted upward in the software stack and built voice agents for enterprise customers. Indian fintech company Cred became the first customer, followed by businesses including Flipkart, Practo, Groww and PolicyBazaar.
Co-founder Siddharth Tripathi said Ringg initially focused on outbound calling, lead qualification and loan collection. Those jobs brought volume but were not sticky, in his assessment, and competition often reduced them to price. The company still supports simpler calls while targeting workflows where completing the task matters more than placing the call.
Ringg currently routes work among different models depending on the use case. It builds its own speech-recognition and generation models and would eventually like to own infrastructure and deployment as well, but Tripathi said the full stack remains too costly today.
Voice becomes one channel in a longer task
The newer workflows include appointment booking for healthcare clinics, abandoned-cart recovery for ecommerce stores and customer onboarding or identity checks for financial applications. Tripathi said Ringg’s agent operates across 1,200 clinics for Practo, helping patients schedule visits and follow up afterward.
Voice calls still account for more than 70% of the business. Ringg is nevertheless adding chat and WhatsApp, and it automates browser-based support requests for clients including Shell. The company wants to present itself as a platform whose agents produce outcomes rather than as a vendor of enterprise phone bots.
That positioning remains a company ambition. The report does not publish completion rates, error rates, customer retention or independent quality comparisons. Healthcare scheduling and financial onboarding also demand consistency; a conversation that sounds natural is not enough if the underlying record or action is wrong.
India anchors expansion and competition
Most customers are in India, with a smaller presence in the Middle East and United States. Rather than selling directly to US businesses, Ringg plans to work with Global Capability Centers in India—the offshore operations hubs used by multinationals—and offer automation alongside human support capacity.
The field is crowded at every layer. Deepgram, ElevenLabs, Cartesia, Sarvam and Smallest.ai develop models; Bolna and Blue Machines also emphasize orchestration; Gnani and Arrowhead focus more heavily on finance. Ringg’s wager is that ownership of the enterprise relationship and completed workflow can be more defensible than a basic voice interface.
Peak XV principal Rishen Kapoor argues that Ringg’s history as a model research lab helps it handle difficult workflows end to end. That is an investor’s assessment, and the source provides no external benchmark confirming it.
Ringg employs 40 people, including more than 15 hired in the past three months. It is recruiting forward-deployed engineers who combine technical and product work, plus researchers tasked with reducing model operating costs. Those roles match its immediate challenge: adapt systems closely enough to finish real business processes while keeping the economics viable.
The funding gives Ringg room to test whether voice AI can become operational software rather than automated calling. Success will depend on outcomes across those expanding channels, not the number of conversations alone.
