Equal scales its Call Screening Assistant to enterprise volume with Sarvam
"(Customer quote: to be updated. A two- to three-paragraph quote from an Equal spokesperson covering the macro problem they set out to solve, why Sarvam was the right partner, and the outcome they have delivered for their customers would mirror the structure of similar Sarvam case studies.)"
(Spokesperson name)
Title , Equal
Introduction
Every day, millions of voice interactions flow through India's financial system, including KYC calls, loan applications, and customer service conversations, in dozens of languages and accents that global AI was never built to understand. Inside that wall of audio sits the question every fintech is now trying to answer: can this be heard, transcribed, and acted on in real time, at scale, in the way Indians actually speak? For Equal, the answer arrived in the form of a Call Screening Assistant built on Sarvam’ STT APIs.
Background
Equal is a digital identity and verification platform that operates at population scale across India, sitting at the centre of how millions of Indians prove who they are and engage with the services that shape their daily lives. Every day, customer service calls, applications, and verification audio flow through Equal's platform in the languages people grew up speaking, often from towns and cities where access to formal services has historically been thin. As that user base scaled, the volume of voice interactions flowing through the platform grew rapidly across Indian languages and regional accents.
Within this context, Equal set out to build a Call Screening Assistant that could process this large and growing pool of voice interactions reliably, in the languages its end users actually speak, and at the latency and uptime standards its enterprise customers demand.
The Problem
Equal's end customers, institutions operating across India, generate voice interactions in a wide range of Indian languages and regional accents. For a Call Screening Assistant to be useful in this environment, the underlying transcription has to perform consistently across that linguistic variety, handle code-switching between English and Indian languages, and remain reliable under high-volume, real-time conditions.
(Specific examples of where global ASR systems fell short for Equal's use case, and the precise pain points the team was looking to solve, would strengthen this section: to be updated.)
The priority was to establish a speech-to-text foundation that could:
- Transcribe voice interactions accurately across multiple Indian languages and regional accents
- Handle code-switching between English and Indian languages reliably
- Maintain accuracy in noisy, real-world audio environments where call quality varies
- Operate at enterprise scale with consistent latency and uptime under load
- Scale elastically as new enterprise clients onboard onto Equal's platform
Why Sarvam
(Equal's reasons for choosing Sarvam over alternatives, including any evaluation or benchmarking process, the alternatives considered, and the specific factors that made Sarvam stand out, would sit here: to be updated.)
What the production usage demonstrates is a strong contextual fit. Equal has progressively upgraded through Sarvam's STT model versions as quality has improved, and now runs saaras:v3 as the primary model behind the majority of its transcription workload. The pace at which Equal has scaled on Sarvam, from around 173,000 calls in October 2025 to a current run rate of approximately 140 million calls per month, reflects sustained confidence in the platform's ability to handle Indian-language audio at enterprise scale.
The Solution
Equal's Call Screening Assistant runs entirely on Sarvam's speech-to-text stack. Across production, the workload is distributed across the full breadth of Sarvam's STT capabilities, including standard transcription, real-time streaming, bulk batch processing, and speaker diarization, with saaras:v3 as the primary model powering the majority of calls.
Transliteration is the dominant mode in production, accounting for roughly 72% of Equal's STT calls. Spoken Indian-language audio is converted into Roman script, which downstream systems in the financial workflow can consume directly. Transcription, translation, code-mix, and verbatim modes are also used where the use case requires them, giving Equal the flexibility to apply the right STT configuration to the right interaction.
(A concrete example of how the Call Screening Assistant works inside a typical call, from incoming audio to actioned output, would help readers visualise the workflow: to be updated.)
Equal has progressively adopted successive Sarvam STT releases as they have shipped, moving through model versions in lockstep with the quality improvements Sarvam has rolled out. That trajectory reflects a partnership where Equal's scale on the platform has grown alongside the maturity of Sarvam's STT models.
The Impact
- Scale: Approximately 4 to 5 million STT calls per day, processing in the region of 34 million audio minutes per day, which translates to roughly 3.4 million audio files per day at an average of about 10 minutes each.
- Growth: Production volume on Sarvam STT has grown approximately 809x from October 2025 to May 2026, moving from around 173,000 calls in October to a current run rate of approximately 140 million calls per month.
- Month-on-month trajectory: Following the production scale-up in October 2025, month-on-month growth ran consistently at triple-digit rates through early 2026 (113% in November, 264% in December, 176% in January, peaking at 608% in February 2026 and 240% in March). Growth has since matured into a stable, high-volume run rate, with month-on-month expansion easing to 38% in April and 14% in May as the platform settled.
- Reliability: 99.98% success rate across approximately 129 million STT calls in the most recent 30-day window.
- Model adoption: saaras:v3 powers approximately 85% of Equal's STT calls, reflecting continuous upgrades through Sarvam's STT model versions as quality has improved.
Looking Ahead
(Equal's forward-looking plans for its use of Sarvam, including any new languages, dialects, endpoints, or use cases on the roadmap, and any additional API capabilities being explored, would sit here: to be updated.)
What the partnership has already demonstrated is that Indian-language voice AI can operate dependably at a scale that very few API customers in India have reached. As Equal continues to expand the reach of its Call Screening Assistant across financial workflows, and as Sarvam continues to advance its multilingual STT capabilities, the collaboration is set to play an increasingly meaningful role in shaping how voice AI is deployed across the Indian fintech sector.