What AssemblyAI shipped in April 2026

Pricing change detected for Assemblyai

Pricing updated for Assemblyai: - Free: allowance changed from $50 credits to 5 streams per minute (scaling to 610 concurrent streams) - Pay-as-you-go: now includes 100 streams per minute (scaling to …

AssemblyAI launches MCP server for AI coding agents

AssemblyAI introduced an MCP server (https://mcp.assemblyai.com/docs) enabling AI coding agents to dynamically fetch and reference its latest API documentation, SDKs, and guides. The server provides t…

Pricing updated for Assemblyai: - Free: allowance changed from Limited usage to $50 credits

AssemblyAI launches Voice Agent API with real-time streaming and tool calling

AssemblyAI introduced a Voice Agent API enabling real-time voice agent creation via a single WebSocket endpoint. The API bundles speech-to-text, LLM inference, text-to-speech, turn detection, and tool…

Speech-to-Speech API is now discontinued

Product was in stock, now marked as discontinued.

Speech-to-Speech API is back in stock

Previously discontinued, now available again.

Pricing updated for Assemblyai: - Free: promotion changed — $50 in credits to use towards AssemblyAI’s Speech-to-Text APIs - Free: now includes Limited usage - New tier: Pay-as-you-go (Custom pricing)

INTRODUCING medical mode

AssemblyAI launched Medical Mode, a medical-grade speech recognition feature reducing missed medical entities (drug names, dosages, diagnoses) by 20% compared to Universal-3 Pro. It achieves lower err…

See what our speech AI can do

AssemblyAI’s Playground page introduces a toolset for testing and demonstrating its speech AI capabilities. Users can upload audio files, apply prompts, and explore features like speaker diarization, …

Speech-to-Speech API is now out of stock

Product was in stock, now marked as out of stock.

Protecting your data is our priority

AssemblyAI emphasizes its commitment to enterprise-grade security by detailing encryption standards (AES-256 at rest, TLS 1.3 in transit), access controls, auditable change management, and compliance …

Powering Voice AI for Industry Leaders

AssemblyAI highlights its Voice AI platform’s adoption by over 5,000 industry-leading companies, including Fortune 500s and startups, demonstrating its scalability and accuracy. Key customers like Cal…

Pricing updated for Assemblyai: - Free: promotion changed — $50 in credits for Speech-to-Text APIs

Zoom leverages AssemblyAI to help advance its AI research and development

Zoom is integrating AssemblyAI’s Speech AI models into its research and development for Zoom AI Companion, a suite of AI-powered collaboration tools. The partnership aims to enhance the accuracy and r…

Universal-3 Pro Streaming Integration with LiveKit Voice Agents

AssemblyAI has integrated its Universal-3 Pro Streaming speech-to-text model into LiveKit’s voice agent framework, enabling real-time transcription with optimized turn detection parameters. The integr…

AssemblyAI Expands Speech Understanding Models with Structured Insights

AssemblyAI has expanded its Speech Understanding models to extract structured insights from audio transcripts, including speaker identification, sentiment analysis, topic detection, and summarization.…

Guardrails: Protect sensitive content in your transcripts

AssemblyAI introduced Guardrails, a suite of features to filter, redact, and moderate transcript content for compliance and safety. The four key capabilities include PII Redaction (automatically detec…

Pricing discovered for Assemblyai

Pricing updated for Assemblyai: - New tier: Free ($0/mo — promo: $50 in credits to use towards AssemblyAI’s Speech-to-Text APIs) - New tier: Pay as you go (Custom pricing)

AssemblyAI Launches LLM Gateway for Unified Access to 20+ AI Models

AssemblyAI launched the LLM Gateway, a unified API enabling access to over 20 large language models (LLMs) from providers like Anthropic, OpenAI, Google, Alibaba Cloud, and Moonshot AI through a singl…

AssemblyAI Expands API with LLM Gateway and EU Endpoints

AssemblyAI has updated its API documentation to consolidate and expand its speech-to-text (STT) and large language model (LLM) gateway services. The most significant change is the introduction of the …

Transcribe a pre-recorded audio file

AssemblyAI has updated its pre-recorded audio transcription documentation to require explicit specification of speech models in every API request, removing any default model. The change affects develo…

Transcribe streaming audio

AssemblyAI has introduced a new streaming transcription capability for real-time audio processing, enabling developers to transcribe live microphone input with sub-300ms latency. The feature requires …

AssemblyAI Launches Medical Mode for Enhanced Medical Terminology Transcription

AssemblyAI has launched Medical Mode, an add-on feature for its pre-recorded speech-to-text (STT) transcription service that significantly improves accuracy for medical terminology. The feature is des…

Universal-3 Pro Streaming Launched by AssemblyAI

AssemblyAI has launched Universal-3 Pro Streaming, a new real-time transcription model optimized for short audio utterances under 10 seconds, featuring sub-300ms latency and native multilingual code s…

Introducing Universal-3 Pro

AssemblyAI introduced Universal-3 Pro, its most advanced Voice AI model, outperforming all existing ASR models in accuracy, particularly for entities and rare words. The model supports both pre-record…

AssemblyAI launched the LLM Gateway, a unified API that provides access to over 20 large language models (LLMs) from providers like Anthropic, OpenAI, Google, Alibaba Cloud, and Moonshot AI through a

AssemblyAI Introduces LLM Gateway for Unified LLM Access

AssemblyAI launched the LLM Gateway, a unified API that consolidates access to over 20 large language models (LLMs) from providers like Claude, GPT, and Gemini into a single interface. This simplifies…

INTRODUCING UNIVERSAL-2

AssemblyAI has launched Universal-2, its most advanced speech-to-text model, delivering significant improvements in accuracy and detail recognition for audio data. The model achieves a 24% improvement…

Earmark achieves 83% cost reduction and unlimited scalability with AssemblyAI

Earmark, a product management AI startup, reduced streaming costs by 83% and achieved unlimited scalability by partnering with AssemblyAI for real-time meeting transcription. The integration enabled E…

How Calabrio boosted customer satisfaction by 80% and accelerated global expansion with AssemblyAI

Calabrio, a workforce and conversation intelligence provider, replaced its stagnant legacy transcription provider with AssemblyAI, achieving an 80% boost in customer satisfaction and accelerating glob…