Rime
Rime Labs is a San Francisco-based voice AI company that builds proprietary enterprise text-to-speech models (Coda, Arcana, Mist) for live customer conversations in healthcare, food ordering, finance, and telecom, sold via usage-based API and on-prem enterprise contracts.
- Company typePrivate
- Founded2022
- HeadquartersSan Francisco, United States
- Headcount11–50
- GTM typeB2B
- OfferingSoftware
What Rime does
Rime Labs, Inc. is a U.S.-based voice AI company that develops and sells enterprise-grade text-to-speech (TTS) models purpose-built for live customer conversations rather than content creation. Founded in 2022 in San Francisco by Lily Clifford, Brooke Larson, and Ares Geovanos, the company operates with a small team of 1-10 employees and is led by CEO Lily Clifford.
Its product portfolio comprises three proprietary TTS model families: Coda (flagship, using a novel dual-decoder autoregressive architecture with a semantic decoder and a 15-code acoustic decoder over 80 ms frames, served by Rime's proprietary TIGERSTRIPE inference stack), Arcana (the most expressive line, with 300+ voices and English/Spanish code-switching), and Mist v3 (a high-throughput engine delivering ~40 ms TTFB p90 on L40S / RTX 6000 GPUs, with a sub-100 ms on-prem variant). All three are built on a proprietary full-duplex conversational speech dataset captured in an in-house San Francisco recording studio and are deployable via cloud API, private VPC, or on-premises, with SOC 2 Type 2 and HIPAA compliance.
Rime monetizes through a usage-based API ($0.03-$0.05 per 1,000 characters), a Growth tier with a $5,000/year minimum, and custom enterprise contracts, distributed via a self-serve developer motion (free credits, public Slack, Rime CLI, docs.rime.ai) and a direct enterprise sales motion. Customers span healthcare (Attune, thymia), food ordering/QSR (Domino's, ConverseNow), telecom (Dialog Axiata PLC in Sri Lanka), voice-agent platforms (LiveKit, Trillet AI, SigmaMind AI), and a Fortune 500 device protection and insurance firm. The company has raised approximately $8.6 million across a 2023 SEC-registered offering and a 2025 round led by Unusual Ventures.
Rime firmographics
Firmographics- Name
- Rime
- Legal name
- Rime Labs, Inc.
- Website
- https://rime.ai
- Company type
- Private
- Founded year
- 2022
- Operating status
- Operating
- Headcount range
- 11–50 employees
- Short description
- Rime Labs is a San Francisco-based voice AI company that builds proprietary enterprise text-to-speech models (Coda, Arcana, Mist) for live customer conversations in healthcare, food ordering, finance, and telecom, sold via usage-based API and on-prem enterprise contracts.
- Ownership category
- akta.pro rank
Rime industry classification
Industry- Product category
- Text-to-Speech Software
- NAICS
- Telephone Apparatus Manufacturing (334210), All Other Telecommunications (517810), Telephone Apparatus Manufacturing (33421)
- SIC
- Services-Computer Programming Services (7371), Telephone Communications (No Radiotelephone) (4813)
- akta.pro primary industry
- Text-to-Speech (TTS) & Voice Synthesis (HDAAAFAB)
- akta.pro secondary industries
- Conversational AI Platforms (chat/voice bots, orchestration) (HDAAAFAC), Contact Center Speech AI (agent assist, IVR automation) (HDAAAFAL), Voice Cloning, Dubbing & Audio Generative Media (HDAAAFAK), Audio & Speech Analytics (call analytics, QA, insights) (HDAAAFAH)
Keywords
Where Rime is headquartered
LocationHeadquarters
- HQ city
- San Francisco
- HQ country
- United States
- HQ region
- North America
Offices2 records
Markets served
Rime business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Technology or R&D, Personnel, Infrastructure, Marketing or Sales, Operations
Revenue model
- Usage-based TTS API pricing: Pay-as-you-go consumption pricing per 1,000 characters generated. Coda: $0.05/1K characters; Arcana: $0.04/1K characters; Mist: $0.03/1K characters. Starter plan includes 3,000 free minutes, 20 concurrent TTS generations, and public Slack support; usage beyond plan limits is billed in arrears. Sign-up promotions include $25 free credits for Coda and $100 in credits.
- Enterprise custom contracts: Custom volume pricing for large organizations running at scale. Includes up to unlimited concurrent TTS generations, unlimited custom TTS voice clones, SLAs and dedicated support, cloud/on-prem/VPC deployment, and BAA (HIPAA) and SOC 2 reports. Long-term / multi-year commitments and on-prem licensing are negotiated directly with the sales team.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Usage-based | Pay-as-you-go | Starter — 3,000 free minutes, then usage-based; 20 concurrent TTS generations; public Slack support |
| Subscription | Multi-year contract | Enterprise — Custom pricing for large organizations running at scale |
| Freemium | Pay-as-you-go | Free tier / promotional credits |
Go-to-market motion3 records
Distribution channels6 records
Marketing channels9 records
Rime product offering
Product offeringCore offering
Rime develops enterprise-grade text-to-speech (TTS) models — flagship Coda, Arcana (multilingual/code-switching), and Mist v3 (low-latency) — designed for live conversational voice applications such as contact center IVR/IVA, voice agents, and phone ordering. Models are delivered through a public TTS API, an on-premises or VPC deployment option, and a developer CLI/dashboard, and are trained on a proprietary full-duplex conversational speech dataset captured in Rime's in-house San Francisco recording studio.
Differentiator
Problem solved
Functional benefit
Brands
- Coda: Text-to-speech (TTS) product/model offering.
- Arcana
- Mist
Products and services
- Coda TTS Model Rime's flagship text-to-speech model designed for high-stakes live enterprise conversations. Coda uses a dual-decoder autoregressive architecture (one semantic decoder, one acoustic decoder over 15 acoustic codes at 80 ms frames) and is served via the custom TIGERSTRIPE inference stack, delivering human-quality voice with sub-700ms end-to-end latency, SpeechQA pronunciation control, and 180+ voices across Professional, Formal, Casual, and Energetic registers. Available via public API, on-prem, and VPC deployment.
- Arcana TTS Model Rime's most expressive TTS model, offering 300+ voices and code-switching support between English and Spanish. Arcana is positioned for expressive enterprise voice experiences and is available through the Rime TTS API alongside Coda and Mist, with usage-based pricing per 1,000 characters.
- Mist v3 TTS Model Rime's low-latency, high-throughput TTS model designed for production enterprise voice agents. Mist v3 delivers ~40ms time-to-first-byte p90 on L40S and RTX 6000 GPUs, sub-200ms cloud latency, <100ms on-prem latency, SSML features including controllable pauses and inline speed adjustment, and Mist's deterministic pronunciation control. Used in production by Attune, LiveKit, Trillet AI, and SigmaMind AI.
- Arcana Sinhala TTS Model A localized Sinhala text-to-speech model co-developed with Dialog Axiata PLC — Sri Lanka's first enterprise-ready Sinhala TTS. Delivered as part of the Arcana model line with on-prem and cloud deployment options for local Sri Lankan enterprises, supporting local-language customer service, IVR, and multilingual voice experiences.
- Rime CLI A command-line developer tool for Rime's TTS platform that lets developers synthesize speech, preview voices, and generate production-ready curl API calls from the terminal. Installable via a one-line shell script, supports 94+ voices, and is usable from agentic coding workflows such as Claude Code, providing a terminal-native entry point to the Rime TTS API.
- Rime TTS API The public, cloud-hosted text-to-speech API that allows developers and enterprises to synthesize speech using Rime's Coda, Arcana, and Mist models, select from 180+ voices, control pronunciation via SpeechQA, and integrate with their own STT and LLM components. Authentication is via API key, with usage-based per-character pricing, public dashboard sign-up, and an SLA-backed enterprise tier.
Quantifiable outcome
- 42%+ call containment rate for a Fortune 500 client
- +9 more outcomes
Companies that use Rime
Customer profileNamed customers9 records
Segments6 records
Ideal customer profiles6 records
Rime technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
AI capability3 records
Feature8 records
Rime partnerships and signals
Strategic signalPartnerships
Ten partnerships are on record, tiered flagship strategic co-development, strategic evaluation partner, evaluation partner, compliance partner, core infrastructure, supporting infrastructure and strategic customer / integration partner.
- Dialog Axiata PLCflagship strategic co-developmentRime Labs and Dialog Axiata PLC (Sri Lanka's #1 connectivity provider) jointly developed and launched Sri Lanka's first enterprise-ready Sinhala text-to-speech model. Rime engineered the full TTS model lifecycle (data pipeline, training data curation, training, evaluation, production deployment) while Dialog contributed domain knowledge, Sinhala language expertise, quality testing, and deployment context. The model is live and ready for deployment across local enterprises.
- Podonosstrategic evaluation partnerAudio evaluation platform purpose-built for voice AI work. Rime partnered with Podonos to run more than 8,000 pairwise listener judgments across the Coda voice pool to rank-order voices by style (professional, formal, casual, energetic). Podonos handles English fluency pre-screening, device/headphone checks, attention tests, noise detection, intelligibility tests, and audio-level normalization.
- Rapidataevaluation partnerAI feedback platform used by Rime for several other evaluations. Rime has been tested against leading TTS providers including ElevenLabs, Google Chirp, and Cartesia on Rapidata's panel of listeners reached through mobile ads, winning 61% / 64% of pairwise judgments.
- Vantacompliance partnerTrust management platform used by Rime to drive its HIPAA requirements analysis and internal audit. Vanta verified over 200 requirements across over 120 controls and is also supporting Rime's SOC 2 Type II audit.
- Amazon Web Services (AWS)core infrastructureCloud hosting and infrastructure sub-processor; data location United States. Powers Rime's cloud deployment of Coda, Arcana, and Mist models.
- Basetencore infrastructureModel inference sub-processor; data location United States. Runs Rime's production model inference workloads.
- Stripecore infrastructurePayment processing sub-processor; data location United States. Rime does not store payment card details directly — billing is handled through Stripe.
- Vercelsupporting infrastructureWeb hosting and deployment sub-processor; data location United States. Hosts the rime.ai marketing site and dashboard front-end.
- Google Cloudsupporting infrastructureSub-processor for development environments; data location United States.
- LiveKitstrategic customer / integration partnerReal-time voice agent platform that deploys Rime's Mist v3 in production. Tom Shapland of LiveKit publicly endorses Mist v3 for voice agents, citing speed and deterministic pronunciation.
Scale indicators10 records
Recent moves7 records
Expansion highlights6 records
Rime competitors and assessment
Company assessmentDirect peers
- ElevenLabs: ElevenLabs is the most-cited direct competitor to Rime, offering high-quality conversational TTS via API and a voice cloning platform. Rime's marketing explicitly positions against ElevenLabs on latency, multilingual code-switching, and enterprise compliance, and the Podonos benchmark measures Rime head-to-head against ElevenLabs.
- Cartesia: Cartesia builds real-time, low-latency TTS and voice AI models used in production voice agents. It is named alongside ElevenLabs and Rime in independent benchmarks and shares Rime's focus on streaming, conversational TTS for agent workloads.
- PlayHT: PlayHT is a developer-focused TTS platform offering voice cloning, multilingual voices, and a streaming API for real-time applications. It overlaps directly with Rime's Coda/Arcana/Mist model lineup and PLG self-serve motion.
- Resemble AI: Resemble AI offers voice cloning and TTS with enterprise controls including on-prem deployment. It competes with Rime in regulated verticals and security-conscious deployments where data residency and custom voice creation matter.
- WellSaid Labs: WellSaid Labs provides studio-quality AI voice for enterprise content and conversational applications. It overlaps with Rime in enterprise voice use cases and brand-grade voice experiences, though it is more weighted toward content production.
- Murf.ai: Murf.ai is an AI voice generator platform serving enterprise and prosumer content use cases with a large voice library. It is comparable to Rime on voice library breadth and developer/enterprise GTM, though it is less focused on real-time conversational TTS.
Broad incumbents
- Microsoft Azure Speech (Azure TTS): Azure's TTS service is a broad incumbent offering inside the Microsoft Cloud stack. It competes with Rime on enterprise compliance (SOC, HIPAA), deployment breadth, and pricing — but as part of a wider speech portfolio rather than a conversational TTS specialist.
- Google Cloud Text-to-Speech: Google Cloud TTS (including Chirp and newer Gemini-powered voices) is named in Rime's Rapidata benchmark comparisons. As a hyperscaler incumbent, it competes on language coverage, scale, and bundled cloud economics, though it is not narrowly focused on conversational enterprise TTS.
- Amazon Polly: Amazon Polly is AWS's general-purpose TTS service and a likely default for many enterprise voice workloads. It competes with Rime on price, AWS-native integration, and breadth of languages — and is named as an infrastructure counterpart via AWS hosting Rime's cloud deployment.
- Speechify: Speechify provides AI TTS across consumer reading apps and enterprise voice tools. It is comparable as a broad TTS provider with both consumer and enterprise distribution, though its strength is content consumption rather than live enterprise conversation.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat5 records
Key risks5 records
Key highlights7 records
Customer concentration
Rime social profiles
Digital presenceRime financial estimates
Financial estimateRevenue estimate
Valuation estimate
Rime leadership team
Management profileNumber of profiles
Profiles3 records
Rime funding detail
Funding detailFunding overview
Funding rounds3 records
Investors7 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Rime M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Rime
What does Rime do?
Rime develops enterprise-grade text-to-speech (TTS) models — flagship Coda, Arcana (multilingual/code-switching), and Mist v3 (low-latency) — designed for live conversational voice applications such as contact center IVR/IVA, voice agents, and phone ordering. Models are delivered through a public TTS API, an on-premises or VPC deployment option, and a developer CLI/dashboard, and are trained on a proprietary full-duplex conversational speech dataset captured in Rime's in-house San Francisco recording studio.
Is Rime a public or private company?
Rime is a private company. It is classified as venture growth investor backed and is currently operating.
When was Rime founded?
Rime was founded in 2022. It employs 11 to 50 people.
Where is Rime based?
Rime is headquartered in San Francisco, United States, in the North America region.
How does Rime make money?
Two revenue lines are on record. Usage-based TTS API pricing is the primary driver. The others are enterprise custom contracts.
Who are Rime's main competitors?
Direct peers on record are ElevenLabs, Cartesia, PlayHT, Resemble AI, WellSaid Labs and Murf.ai. Broad incumbents are Microsoft Azure Speech (Azure TTS), Google Cloud Text-to-Speech, Amazon Polly and Speechify.
Does Rime have an API?
Yes. Rime offers a public, cloud-hosted text-to-speech API (with on-prem and VPC deployment options) that allows developers to synthesize speech using the Coda, Arcana, and Mist models, select from 180+ voices, control pronunciation via SpeechQA, and integrate with their own STT and LLM components. The API includes a "Build a curl command" feature generated by the Rime CLI, with public dashboard sign-up and per-character pricing. Authentication is via an API key saved to ~/.rime/rime.toml. Developer documentation is at docs.rime.ai/api-reference/quickstart.
What industry is Rime in?
Rime's product category is Text-to-Speech Software. Its primary akta.pro industry code is HDAAAFAB, Text-to-Speech (TTS) & Voice Synthesis, with a secondary code of HDAAAFAC, Conversational AI Platforms (chat/voice bots, orchestration). Its NAICS code is 334210 and its SIC code is 7371.