AssemblyAI publishes JavaScript tutorial for browser voice assistant
Blockchain.News reported that AssemblyAI published a JavaScript tutorial on Oct. 13, 2026, for building a browser voice assistant over one WebSocket.
Why it matters
This is vendor documentation from an AI speech company, not a crypto development: no token, protocol or listed asset is involved, and Blockchain.News carried it as general tech news. Its significance for crypto markets is limited, and the only evidence in this set is a single report restating AssemblyAI's own claims.
Key facts
AssemblyAI published a JavaScript tutorial on Oct. 13, 2026 showing how to build a browser voice assistant that listens through the microphone, sends audio over one WebSocket and plays replies back in the tab.
Reported by Blockchain.News
AssemblyAI says the browser's Web Speech API handles speech recognition and browser text-to-speech but leaves understanding logic, interruption handling and LLM integration to the developer.
Reported by Blockchain.News
AssemblyAI says its Voice Agent API runs speech recognition, LLM orchestration and text-to-speech server-side behind one WebSocket, leaving JavaScript to capture audio bytes and render transcripts.
Reported by Blockchain.News
The browser client connects to wss://agents.assemblyai.com/v1/ws using a temporary token minted by a small Node server, keeping the API key on the server.
Reported by Blockchain.News
The browser sends 24 kHz, 16-bit PCM audio to the endpoint and receives 24 kHz PCM audio back.
Reported by Blockchain.News
The example captures the microphone with an AudioWorklet and resamples audio inside the worklet so the code behaves consistently across Chrome, Firefox and Safari.
Reported by Blockchain.News
AssemblyAI says the flow is designed to reduce latency and simplify the pipeline by avoiding separate speech-to-text, LLM and text-to-speech vendors; the report gives no independent measurement.
Reported by Blockchain.News
The tutorial presents the browser voice assistant as suited to conversational assistants rather than simple command-and-response voice controls.
Reported by Blockchain.News
What happened
Blockchain.News reported on Oct. 13, 2026 that AssemblyAI published a JavaScript tutorial showing how to build a browser voice assistant that captures microphone audio, sends it over a single WebSocket and plays replies back in the tab. The tutorial sets the company's Voice Agent API against the browser's built-in Web Speech API: AssemblyAI says the Web Speech API handles speech recognition and browser text-to-speech but leaves understanding logic, interruption handling and LLM integration to the developer, while the Voice Agent API runs speech recognition, LLM orchestration and text-to-speech server-side, leaving JavaScript to capture audio bytes and render transcripts. Per the tutorial, the browser connects to wss://agents.assemblyai.com/v1/ws with a temporary token minted by a small Node server so the API key stays on the server, and the browser sends 24 kHz, 16-bit PCM audio while receiving 24 kHz PCM back. The example captures the microphone with an AudioWorklet and resamples audio inside the worklet so the code behaves consistently across Chrome, Firefox and Safari. AssemblyAI says the single-vendor flow is designed to cut latency and simplify the pipeline by avoiding separate speech-to-text, LLM and text-to-speech vendors; that is the company's own claim, and no second outlet here tested it. The article text available ends mid-sentence while describing how reply audio chunks are scheduled for playback, so that part of the report is incomplete.
How the story developed
- First report
First report by Blockchain.News
AssemblyAI outlines browser voice assistant build in JavaScript
Affected entities
Original sources
| Publisher | Report | Role | Published |
|---|---|---|---|
| Blockchain.News Crypto media | AssemblyAI outlines browser voice assistant build in JavaScript | Original |