Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.
Speech-to-text for AI medical scribes: Why clinical vocabulary breaks generic STT
TL;DR: Generic STT engines fail in clinical environments because language model probability overrides correct acoustic detection of medical terms, substituting phonetically plausible but clinically wrong candidates silently. The result corrupts drug names, dosages, and diagnoses before the LLM ever sees them. Before selecting an STT engine for a medical scribe, verify four things: whether vocabulary biasing works at inference time without fine-tuning, whether async diarization accurately separates clinician and patient audio, whether the model holds up on noisy consultation recordings rather than clean read-speech, and whether the vendor's data training policy covers PHI by default on your plan.
Migrating from self-hosted Whisper to a managed speech-to-text API
TL;DR: Self-hosting Whisper's true cost rarely sits in the model weights. GPU idle time, VRAM leaks under parallel load, and the engineering hours spent maintaining CUDA dependencies and diarization pipelines are where the bill compounds. For teams processing under roughly 3,000 hours per month, assuming 20% of one US FTE at $150K loaded annual cost, a managed API is cheaper, though the break-even shifts materially against your actual labor cost. Above that threshold, the decision depends on your DevOps overhead and whether audio accuracy on real-world recordings matters for downstream systems like CRM sync and coaching scores.
Migrating from AssemblyAI to Gladia: A step-by-step switching guide
TL;DR: Switching from AssemblyAI requires four concrete changes: update one auth header, remap batch endpoints, adjust the JSON response schema, and resample audio for WebSocket connections. Multiple customers independently report completing these in under a day with a rollback abstraction layer in place. The bigger structural difference is cost model: a production stack with diarization, sentiment, entities, and summarization runs $0.30/hr on AssemblyAI's Universal-2 tier because each feature is metered separately, versus a bundled base rate. This guide covers the exact parameter mappings, payload diffs, WebSocket reconfiguration, and a zero-downtime cutover strategy.
How Aircall cut transcription time by 95% with Gladia
Published on Oct 9, 2025
The contact center is transforming. Traditionally defined by manual workflows, siloed data, and reactive customer service, today's Contact Center as a Service (CCaaS) platforms are embracing a new era—one driven by real-time AI and automation.
Transcription lies at the core of this transformation. Converting voice to text with speed and precision unlocks a cascade of next-gen capabilities: automated summaries, sentiment detection, agent coaching, CRM enrichment, and more. But many legacy or in-house solutions fall short—too slow, too inaccurate, or too resource-heavy to scale.
Aircall, the leading AI-powered voice platform for growing businesses, recognized this inflection point early. To meet the growing demand for fast, intelligent insights from customer conversations, Aircall turned to Gladia’s speech-to-text API.
Here’s how Aircall reduced transcription time by 95%, empowered its users with near-instant insights, and laid the groundwork for a smarter, AI-driven CCaaS future.
About Aircall
Aircall is an integrated customer communications and intelligence platform. It unifies voice and digital channels into one seamless platform, offering one-click integrations with leading CRMs and over 250 business tools. With a strong focus on cloud-based voice solutions, Aircall helps teams streamline conversations, improve customer support, and drive sales efficiency.
Farid Issabhai, Staff Engineer at Aircall, is at the forefront of Aircall’s AI and transcription initiatives. He played a key role in integrating cutting-edge technologies, including Gladia’s speech-to-text API, into Aircall’s workflows.
Challenge: More accurate, fast, and scalable transcription for global telephony
As a leading voice platform, Aircall processes thousands of calls every day across diverse languages and use cases, from customer support to sales interactions. Initially, Aircall developed an in-house transcription engine, but maintaining and improving it proved challenging.
Solution: Gladia’s speech-to-text API
After evaluating different STT API vendors, Aircall chose Gladia for its strong performance in transcription accuracy, especially for key strategic languages.
Gladia’s API allowed Aircall to:
✓ Transcribe calls across multiple languages like Spanish, German, and Italian.
✓ Process over 1M transcriptions per week
✓ Deliver transcripts significantly faster than their previous solutions.
How Aircall uses transcription
Aircall integrates Gladia’s transcriptions as a foundational layer for advanced features for their CCaaS platform:
Searchability: Users can search for keywords across calls
AI-generated insights: Summaries, key topics, and sentiment analysis are built on top of the transcripts
Agent coaching: Aircall’s coaching features assess calls for compliance and training, evaluating factors like greetings or responses to objections
CRM Integration: While transcriptions aren’t logged directly into CRMs like HubSpot or Salesforce, summaries and AI insights are pushed via webhooks
Farid explains,
Why Aircall chose Gladia
Aircall’s decision to partner with Gladia was driven by:
Accuracy: High performance across key languages benchmarked on internal datasets composed of phone call audio
Speed: Drastically reduced transcription delays
Developer Experience: A well-designed API that simplified integration
Cost-Effectiveness: A solution that balances performance with the economics of scaling
Results: Faster insights, smoother operations
Since switching to Gladia:
Transcription times have dropped from up to 30 minutes to under 1.5 minutes
Aircall processes around 1M calls weekly, enabling scalable AI features
Improved user satisfaction by delivering faster insights
Farid highlights,
Looking ahead
Aircall is exploring new frontiers with real-time transcription and AI voice agents. While asynchronous transcription currently meets most needs, the team is actively experimenting with new features like real-time assistance during sales calls, where AI can suggest responses based on conversation context.
Farid shares,
Final thoughts
Farid’s advice for companies looking to integrate speech-to-text AI:
About Gladia
Gladia provides a speech-to-text and audio intelligence API for building virtual meeting and note-taking apps, call center platforms, and media products, providing transcription, translation, and insights powered by best-in-class ASR, LLMs, and GenAI models.
After reading this case study, do you think Gladia could be the right fit for your business?
