November 17, 2026
What Is Text to Speech (TTS)? How It Works for Developers
Text to speech (TTS) turns written text into a spoken voice. Learn how neural TTS models work, how fast they start speaking and how TTS APIs bill.
Building real-time voice AI on the web: architecture, providers, latency and data sovereignty.
November 17, 2026
Text to speech (TTS) turns written text into a spoken voice. Learn how neural TTS models work, how fast they start speaking and how TTS APIs bill.
November 16, 2026
We rank eight decisions APIs, led by Gemini Decisions, Jev, OpenAI Decisions and Clef, on accuracy, latency, price, inputs and data terms.
November 12, 2026
Mozilla archived DeepSpeech in 2025 and its fork Coqui STT went unmaintained in 2023. Pick the open model for each DeepSpeech use, Node code included.
November 10, 2026
Automatic speech recognition turns speech into text. Learn how ASR models work, what word error rate hides and what a live voice agent needs from one.
November 8, 2026
Eight voice agent testing platforms ranked on what a web app agent needs: simulated callers without a phone number, regression tests in CI and call monitoring.
November 5, 2026
This ranking compares ten open source speech to text models for voice agents on streaming, word error rate, CPU or GPU, languages and a license you can ship.
November 3, 2026
LiveKit is free to self-host. LiveKit Cloud bills $0.01 an agent minute past a free tier. A web agent costs about $0.055 a minute, providers included.
November 1, 2026
Compare Mistral’s Voice Agents SDK, a custom WebSocket pipeline and Micdrop. Then build a Node voice agent with Voxtral 3 for transcription and speech.
October 31, 2026
Browser echo cancellation reduces speaker audio in microphone input. Learn how getUserMedia requests it, check settings and handle remaining echo.
October 31, 2026
Build a multilingual voice agent by choosing a language per call, detecting it from speech, or supporting code switching, then matching the LLM and voice.
October 31, 2026
Compare four local LLM servers for an app by API coverage, tool calls, latency, concurrency, model formats and deployment, then choose the right fit.
October 31, 2026
Choose SSE for server-to-browser AI text streams and WebSocket for two-way audio or live control. Compare direction, reconnects, binary data and operations.
October 31, 2026
A voice agent needs to detect speech during playback, stop every output path, cancel stale generation and keep only the answer the user heard.
October 30, 2026
A voice agent can speak before its LLM finishes by streaming text into TTS and audio out as it is made. Here is how to choose, buffer and measure that path.
October 30, 2026
Vapi gives developers more provider choice. Retell AI brings phone workflows and call-center tools. Compare pricing, APIs, data controls and when to self-host.
October 29, 2026
The speechSynthesis API is free and works offline, but each device speaks with its own voices. A server voice or a model in the tab sounds the same everywhere.
October 28, 2026
Pocket TTS clones a voice on a CPU and Qwen3-TTS on a GPU, Phonon from a voice file. Sample lengths, licenses, latency measured from Node, and the code.
October 28, 2026
Pipecat is a Python pipeline with your choice of transport. LiveKit Agents is a WebRTC room in Python and Node. How to choose, with dated facts.
October 27, 2026
This ranking compares ten open source text to speech models for voice agents on first audio delay, CPU fit, languages, voice cloning and license.
October 23, 2026
Bland AI bills $0.11 to $0.14 a minute by plan, models and voice included. We price each plan and fee, then the same call on providers you pay directly.
October 23, 2026
To start a voice agent hands-free in the browser, compare openWakeWord, Porcupine Web and a transcript regex on licenses, costs, code and false triggers.
October 22, 2026
In 2026, VAD and turn detection run on the user’s device. Speech to text, the LLM and the voice can too, but run better on a server you operate.
October 21, 2026
We compare ElevenLabs, Cartesia, Gradium and Voxtral with Phonon, Piper and Qwen3-TTS running in Node, on French latency and licenses.
October 19, 2026
Vapi charges $0.05 a minute, then each provider bills on top. We add up a typical minute, the plans and concurrency, then price the same stack on your server.
October 18, 2026
Retell AI bills $0.07 to $0.31 a minute. We add up each line, the add-ons and concurrency, then compare the bill with running the same providers yourself.
October 17, 2026
Nine LLMs for voice agents, ranked on a multi-turn voice benchmark and compared on tool calls, time to the first word, price, EU processing and open weights.
October 17, 2026
Eight speech to speech APIs ranked for voice agents on the Artificial Analysis benchmark, compared on price per minute, latency, tools and EU processing.
October 16, 2026
We ranked ten streaming speech recognition APIs for voice agents on accuracy, time to the final transcript, turn detection, languages and price per minute.
October 14, 2026
Browsers record at 44.1 or 48 kHz, speech models read 16 kHz and many voices speak at 24 kHz. Learn the rate each stage expects and what resampling costs.
October 13, 2026
gpt-6-luna, Jev and Claude Haiku 4.5 read the same 90 voice commands. Compare their latency, the commands they got right and the price of 1,000 turns.
October 12, 2026
Call the OpenAI Decisions API from Node, get typed answers from gpt-6-luna in 200 to 530 ms, and route each turn of a voice call before the LLM starts.
October 11, 2026
expo-speech and react-native-tts read text with the voice of the phone. A voice agent streams a provider voice from a server. Choose one, then build it.
October 10, 2026
Vocode has shipped no release since 2024. Micdrop is a maintained open source alternative for a voice agent in a web app, with client and server in TypeScript.
October 9, 2026
Bland runs its own models and bills $0.11 to $0.14 a minute. Micdrop runs the voice loop in your Node server, on the providers and keys you choose.
October 9, 2026
WebRTC suits weak networks, group calls and phone lines. For one browser talking to your own server, a WebSocket carries the call with far less to run.
October 7, 2026
Ollama, Whisper and Piper make a voice agent that answers in about a second on a laptop, with no API key. We cover the setup, the latency and the traps we hit.
October 6, 2026
How microphone audio becomes live text over one WebSocket in Node: voice detection in the browser, 16 kHz PCM chunks, five messages and a streaming STT.
October 5, 2026
Synthflow now starts at $30,000 a year. Micdrop is an open source TypeScript library that runs the voice loop in your Node server, billed by your providers.
October 4, 2026
Un agent vocal IA qui garde la voix en Europe, avec Mistral, Gladia et Gradium en API ou des modèles open source chez vous. Code TypeScript et limites.
October 3, 2026
A voice agent ends your turn on silence, on the sound of your sentence or on its words. Compare the three, then run Smart Turn in TypeScript.
October 2, 2026
We turn the ElevenLabs API prices of October 2026 into a cost per minute of agent speech and count how many calls you can run at once on each plan.
October 1, 2026
We compare Cartesia Sonic 3.6 with ElevenLabs Flash and v4 Turbo on latency, voices, price and concurrency, then run both in one TypeScript voice agent.
September 30, 2026
Only Eleven v3 and v4 perform ElevenLabs audio tags. In a voice agent, prompt the LLM to write them, show them as labels and strip them for a backup voice.
September 29, 2026
Eleven v4 Turbo streams only over the Text to Dialogue WebSocket. With a flush per sentence, it started speaking in 138 ms in our launch-day test.
September 28, 2026
OpenAI, Gemini, Voxtral and Gladia compared for live transcription in a voice agent: price per minute, languages, streaming accuracy and time to transcript.
September 27, 2026
Keywords, an LLM or a dedicated model can classify each turn of a voice call. Compare their latency and route the answer in TypeScript before the LLM speaks.
September 26, 2026
Jev, the System One model of TypeSafe AI, answers typed questions about a text with calibrated probabilities in 70 to 500 ms. It writes no text.
September 25, 2026
Classify each turn of a voice call with Jev in a few hundred ms, then pick who answers before the LLM starts: a scripted line, a human, plain code or the LLM.
September 24, 2026
Ten voice AI SDKs for web and React Native apps, ranked on TypeScript coverage, React Native support and provider choice. Micdrop, the publisher, is second.
September 22, 2026
The Realtime SDK needs a WebRTC transport, an audio session and an ephemeral key on the phone. Hold the session on your server and the app only streams PCM.
September 18, 2026
GPT-Live-1 listens while it speaks, GPT-Realtime-2.1 reasons and calls tools in one session, and Gemini 3.8 Live costs about a quarter of GPT-Realtime-2.1.
September 15, 2026
Whisper transcribes in Node with no API key but pads every sentence to thirty seconds. Vosk, Moonshine and Parakeet read audio at its real length.
September 13, 2026
This ranking compares ten AI voice agent platforms on price per minute, telephony, provider choice and EU hosting. Micdrop publishes it and sits fourth.
September 12, 2026
Cartesia, Gradium, Mistral, OpenAI, Gemini and four local engines share one TypeScript interface. Switching engine takes one line. FallbackTTS keeps a backup ready.
September 9, 2026
The Web Speech API transcribes in Chrome, Safari and desktop Firefox. Chrome and Safari send your audio to Google or Apple, while Firefox transcribes on the device.
September 2, 2026
Five engines speak from Node with no API key. Phonon speaks first, Piper covers 53 languages, Kokoro installs in one command, Qwen3-TTS needs a GPU.
August 30, 2026
The system recogniser, a model on the device, or streaming to a server. Three ways to transcribe voice in a React Native app, and what each one costs.
August 24, 2026
Micdrop runs the voice loop in your Node server, so you pay your AI providers directly. Retell AI bills it per minute and extra concurrent calls monthly.
August 19, 2026
The Realtime API gives you speech to speech in one connection. A pipeline gives you provider choice, voices and cost control. Here is how to pick between them.
August 16, 2026
Vapi runs your voice agent on its servers for $0.05 a minute. Micdrop runs the same loop inside the Node server you already ship, with no platform fee.
August 13, 2026
Micdrop runs a voice call in TypeScript over a WebSocket to your own Node server. LiveKit Agents works in Node too, as a worker that joins a WebRTC room.
August 13, 2026
Thirteen voice agent frameworks and hosted platforms compared on language, transport, licence and cost. Micdrop publishes this ranking and sits third in it.
August 13, 2026
A browser voice agent needs voice activity detection to tell when the user speaks. Compare volume thresholds, WebRTC VAD and Silero, then tune your choice.
March 3, 2026
Micdrop runs the whole voice AI pipeline in TypeScript, browser and Node.js side, with provider fallback and semantic turn detection built in.
March 3, 2026
Intégrez Gradium, le TTS français issu de Kyutai, dans une application web avec Micdrop. Vos données passent par des serveurs européens, avec un fallback.