Hume API

What it does

Hume API is a set of developer voice APIs from Hume AI. The Empathic Voice Interface (EVI) takes in a user's speech, measures vocal modulations, and responds through a speech-language model that generates both language and speech. Text-to-Speech (TTS), built on the Octave model, takes in text and natural-language descriptions and produces expressive speech audio through streaming or non-streaming endpoints. Voices can be chosen from a library of over 100, designed from descriptive prompts, or cloned from a speech sample with user consent.

Developers use these APIs for interviewing and coaching simulations, digital companions, digital assistants, narration and podcast audio, lessons, and character voices. SDKs for React, TypeScript, Python, Swift, and .NET handle authentication, requests, and streaming, and a no-code Playground lets teams test and configure EVI. EVI emphasizes emotional prosody, end-of-turn detection, and interruptibility, while Octave uses language-model context to shape how words are spoken.

This is aimed at developers and product teams building voice applications rather than end users. Using the TTS and EVI APIs requires an account on Hume's platform and an API key. Both APIs end on November 13, 2026 at 12:01 a.m. EST and remain fully supported until then, after which account data is permanently deleted, so new projects should plan around that date. Hume points to Real World VoiceEQ, an evaluation platform, for comparing alternative voice models.

What it can do

• Real-time emotionally responsive voice interaction through EVI with WebSocket and REST endpoints
• Expressive text-to-speech synthesis through Octave with streaming and non-streaming endpoints
• Voice library of over 100 expressive voices
• Custom voice creation from descriptive prompts (Voice Design)
• Voice cloning from a recorded or uploaded speech sample with user consent
• No-code Playground for testing and configuring EVI

Before you choose

• Both the TTS and EVI APIs end November 13, 2026 at 12:01 a.m. EST, and account data is permanently deleted after November 13, 2026.
• Hume describes VoiceEQ as an evaluation platform for comparing voice models, not a replacement speech API or drop-in substitute.
• Voice cloning requires user consent for the speech sample used.

Listed under

• Coding
• API
• Audio narration
• Developers
• Emotional AI
• Speech-to-speech
• Text-to-speech

Get started

Create an account at the Hume cloud console to get API access before November 13, 2026.

1. Create an account and open the Hume platform at cloud.hume.ai to obtain API access.
2. Test and configure EVI in the no-code Playground before writing code.
3. Install an SDK for your environment (React, TypeScript, Python, Swift, or .NET) and call the EVI WebSocket or TTS endpoints using the API reference.

Confirm the applicable plan, usage charges and commercial terms with the provider.

We couldn’t find published prices for Hume API. See their pricing page ↗

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Welcome to Hume AI | Hume API ↗dev.hume.ai

Company website

dev.hume.ai: product, pricing and policy pages

YC directory mirror

Batch and company name

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.