Hume AI

Name

Hume AI

Legal name

Hume AI, Inc.

Website

https://hume.ai

Company type

Private

Founded year

2021

Operating status

Operating

Headcount range

11–50 employees

Ownership category

akta.pro rank

Product category

Voice AI Platform

NAICS

Custom Computer Programming Services (541511), Software Publishers (5132)

SIC

Services-Prepackaged Software (7372)

akta.pro primary industry

Conversational AI Platforms (chat/voice bots, orchestration) (HDAAAFAC)

akta.pro secondary industries

Contact Center Speech AI (agent assist, IVR automation) (HDAAAFAL), On-Device Speech & Audio AI (wake word, ASR, enhancement) (HDAAAJAF), Voice Assistants & Voice UI (VUI) (HDAAAFAD)

Keywords

• Voice AI platform
• Speech-to-speech models
• Text-to-speech systems
• Emotion recognition API
• Voice AI datasets

Headquarters

HQ city

New York

HQ country

United States

HQ region

North America

Offices3 records

Markets served

GTM type

B2B

Offering type

Software

Cost components

Technology or R&D, Personnel, Infrastructure, Marketing or Sales, Operations

Revenue model

1. API subscriptions and usage-based pricing: Tiered subscriptions (free, Starter, Creator, and Pro/Enterprise tiers ranging from $3 to $500/month) plus pay-as-you-go metered usage of TTS and emotion APIs at $0.0X/min with volume discounts. Recurring SaaS plus consumption-based revenue mix.
2. Pay-as-you-go API consumption: Metered usage of EVI speech-to-speech and Octave TTS endpoints billed per minute/character beyond tier allotments, with volume discounts for high-traffic applications.
3. Enterprise research and data services: Custom research partnerships providing voice datasets, annotations, evaluation tools, and workflows for frontier AI labs and AI-first enterprises; includes multi-year, 8-figure enterprise contracts (multiple signed in January 2026) and projects >$100M in revenue for 2026.
4. Technology licensing royalties: Non-exclusive licensing of Hume's emotional voice recognition and expression-measurement technology to large platforms such as Google DeepMind, plus continued licensing of technology to other AI labs and clients.
5. Freemium self-serve access: Free tier for non-commercial evaluation and prototyping; designed to funnel developers into paid Creator and Enterprise plans.

Pricing tiers

Go-to-market motion1 record

Distribution channels6 records

Marketing channels10 records

Core offering

Hume AI builds voice AI foundation models with embedded emotional intelligence. Its core offerings are the Empathic Voice Interface (EVI) speech-to-speech model, the Octave LLM-based text-to-speech system, the open-source TADA TTS model, and supporting APIs for emotion measurement, human-feedback evaluation, and curated multilingual voice training datasets.

Product overview

Hume AI offers an integrated platform-plus-modules architecture for building voice-first AI applications with emotional intelligence. The platform consists of three primary product pillars: (1) the Empathic Voice Interface (EVI), a closed-source speech-to-speech foundation model that detects 48+ emotional expressions and supports real-time empathic conversations; (2) Octave, a closed-source LLM-based text-to-speech system with voice design, voice cloning, and voice conversion capabilities; and (3) TADA, an open-source TTS system for on-device deployment. The platform is augmented by supporting modules including the Expression Measurement API for emotion detection, Human Feedback API (Study Runner) for human evaluations, Training Data library with curated multilingual speech datasets, and the Hume MCP Server for developer integration. All products are accessible via a unified REST/WebSocket API at api.hume.ai with SDKs for TypeScript, Python, .NET, Swift, and React, and integrations with Twilio, LiveKit, Pipecat, Vapi, Agora, and Vercel AI SDK.

Differentiator

Problem solved

Functional benefit

Brands

• EVI (Empathic Voice Interface): Hume's closed-source LLM speech-to-speech system with interruptibility, backchanneling, expressive instruction following, and external LLM compatibility.
• Octave
• TADA
• OCTAVE (Omni-Capable Text and Voice Engine)

Products and services

• Empathic Voice Interface (EVI)
• Octave
• TADA (Text-Acoustic Dual Alignment)
• Expression Measurement API
• Human Feedback API (Study Runner)
• Training Data / Data Library
• Enterprise Research and Data Services

Named customers9 records

Segments11 records

Ideal customer profiles4 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

Integration17 records

AI capability13 records

Feature8 records

Partnerships

19 partnerships are on record, tiered flagship and core.

• Google DeepMindflagshipStrategic or Co-development Partner · 22 January 2026On January 22, 2026, Google DeepMind signed a non-exclusive licensing agreement with Hume AI; co-founder/CEO Alan Cowen and approximately seven top engineers joined Google to enhance Gemini's voice capabilities. Hume AI continues operating independently under new CEO Andrew Ettinger and continues licensing its technology to other AI firms and clients.
• WebAppCloudscoreImplementation/ SI/ Consulting Partner · 6 November 2025WebAppClouds built a voice AI phone support solution called EVI with Hume to replace traditional IVR menus for multi-location brands, supporting proactive outreach and inbound support across multiple languages.
• GAFcoreTechnology or Integration · 3 November 2025GAF integrated Hume's text-to-speech technology into its training programs and marketing voiceovers to deliver consistent, high-quality audio for thousands of contractors, retailers, and employees.
• AudioStackcoreChannel Partner/ Reseller/ Distributor · 24 October 2025AudioStack integrated Hume's emotionally intelligent voices into its enterprise audio production platform used by Publicis, Omnicom, iHeartMedia, and Dentsu, enabling broadcast-ready ad and branded audio 10x faster at up to 80% lower cost.
• JourneecoreStrategic or Co-development Partner · 23 October 2025Journee integrated Hume's EVI into its "Human AI" platform serving global brands. Hume replaced OpenAI, Deepgram, ElevenLabs, and Langfuse, halving Journee's costs and delivering 140ms–1.3s latency for AI agents serving BMW, Shiseido, Macy's, H&M, and Clinique.
• Inception PointcoreStrategic or Co-development Partner · 10 October 2025Inception Point partnered with Hume to power 50+ AI virtual personalities producing 5,000+ shows and 3,000+ episodes weekly at $1 per episode, reaching 11 million listeners and expanding into short-form video and interactive experiences.
• SambaNovaflagshipTechnology or Integration · 1 October 2025SambaNova and Hume co-developed a custom inference stack enabling Octave 2's sub-200ms latency. SambaNova hosts Hume's EVI 4 Mini and Octave 2 on its SambaCloud platform, supporting 11 languages, real-time translation, and emotional expressiveness for global enterprise deployments.
• CoconotecoreTechnology or Integration · 24 September 2025Coconote integrated Hume EVI to power its voice chat feature, transforming static notes into dynamic conversations for neurodiverse students and professionals. Coconote ranks in the Top 50 Education apps on Apple's App Store.
• Reality DefendercoreTechnology or Integration · 26 August 2025Reality Defender partnered with Hume to enhance voice synthesis threat detection, build datasets, and develop proactive defenses against sophisticated synthetic voice threats for enterprise and government customers.
• Niantic SpatialcoreStrategic or Co-development Partner · 7 April 2025Niantic Spatial, in partnership with Snap Inc. (Spectacles hardware) and Hume AI (voice), built location-aware AI companions blending Niantic's Large Geospatial Model, Snap's AR glasses, and Hume's EVI for natural, emotionally intelligent conversation.
• Snap Inc.coreStrategic or Co-development Partner · 7 April 2025Snap Inc. provides Spectacles AR glasses hardware in a three-way partnership with Hume AI and Niantic Spatial to deliver emotionally intelligent, spatially aware AI companions.
• AnthropicflagshipTechnology or Integration · 22 November 2024Hume AI integrated its Empathic Voice Interface (EVI) with Anthropic's Claude AI models to deliver emotionally intelligent voice interactions. The collaboration has powered over 2 million minutes of AI voice conversations; Claude is also listed as a natively supported supplemental LLM in EVI configurations.
• TwiliocoreTechnology or IntegrationDocumented integration for telephony: Hume EVI supports inbound phone calling via Twilio, with Twilio metadata fields (call_sid, account_sid, from/to numbers, geo data) surfaced through chat webhooks.
• LiveKitcoreTechnology or IntegrationDocumented integration for real-time voice infrastructure; LiveKit is listed as a supported integration for embedding Hume's voice models into live applications.
• VapicoreTechnology or IntegrationDocumented integration enabling Vapi-powered AI agent platforms to use Hume's voice models for emotionally intelligent conversational agents.
• PipecatcoreTechnology or IntegrationDocumented integration with Pipecat's open-source framework for building real-time, multimodal conversational AI agents on Hume's voice models.
• Vercel AI SDKcoreTechnology or IntegrationDocumented integration letting Vercel AI SDK applications directly access Hume's voice models for in-app voice experiences.
• AgoracoreTechnology or IntegrationDocumented integration with Agora's real-time engagement platform for embedding Hume's voice models in live audio applications.
• The Hume InitiativecoreOthersNonprofit founded by Hume AI (thehumeinitiative.org) that develops ethical guidelines for empathic AI. Hume embeds these guidelines into its contractual terms of service as a differentiator for enterprise trust.

Scale indicators13 records

Recent moves6 records

Expansion highlights7 records

Direct peers

• ElevenLabs: Closest direct competitor in LLM-based TTS and voice cloning; Octave has been benchmarked head-to-head against ElevenLabs Voice Design in blind studies. ElevenLabs is the market-leading voice AI startup serving very similar use cases (content creation, localization, agent voice).
• PlayHT: AI voice generation platform with voice cloning and an LLM-based TTS stack, targeting similar content creation, agent, and enterprise use cases as Octave and EVI. Operates as a direct SMB/enterprise alternative to Hume.
• Resemble AI: Voice cloning and TTS provider offering real-time voice synthesis, emotion control, and deepfake detection; overlaps with Hume on expressive TTS, voice cloning, and the Reality Defender-style synthetic voice defense use case.
• WellSaid Labs: Enterprise-focused AI voice platform for training, marketing, and corporate communications — a direct comparator to Hume's GAF/AudioStack use cases. Positions as a studio-quality alternative to ElevenLabs for branded content.
• Speechify: Consumer and enterprise TTS product with voice cloning and a creator marketplace; competes with Octave in content creation and accessibility verticals and shares Hume's freemium-to-enterprise GTM motion.
• Lovo AI: Generative AI voice platform with voice cloning, emotion control, and multilingual TTS; closest peers in feature set to Octave and EVI for marketing, e-learning, and entertainment customers.

Broad incumbents

• OpenAI: Provides the Realtime API and GPT-4o voice capabilities, which Hume positions against in conversational voice. Journee explicitly replaced OpenAI with Hume EVI, indicating overlap but OpenAI is a much broader foundation model platform.
• Google Cloud (Gemini Live / Vertex AI Speech): Operates the Gemini Live voice API and Google Cloud Text-to-Speech, competing directly with EVI and Octave; Google DeepMind is also Hume's licensing partner, making it both a customer and indirect competitor for the same workloads.
• Microsoft Azure AI Speech: Enterprise TTS and conversational AI offering within the Azure cloud platform, providing neural voices, custom voice, and real-time speech services to large enterprises in direct competition with Hume's enterprise voice stack.

Emerging players

• AssemblyAI: Speech-to-text and audio intelligence API provider that has expanded into real-time and voice agent capabilities; competes with the EVI stack in the speech understanding and conversational AI layer for developers.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks6 records

Key highlights7 records

Customer concentration

• GitHubgithub.com/HumeAI
• LinkedInlinkedin.com/company/hume-ai
• Xx.com/hume__ai

Compliance3 records

Revenue estimate

Valuation estimate

Number of profiles

Profiles2 records

Funding overview

Funding rounds3 records

Investors8 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments1 record

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Hume AI