Opportunities › Local text-to-speech with voice cloning › Proposed app
Hearthvoice
Text to speech and voice cloning that runs on your own device
- Score
- 58
- confidence 55
- Evidence
- 9
- items on the page
- Rivals
- 2
- named in sources
- First version
- 45 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2023
- Emerging market · AI estimate
ScreensConcept
Desktop app: Content creators write scripts, tune emotional tone per line, and batch-render files—tasks requiring detailed editing and sustained creative work. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
Script
Ready to render- Line 1 · NarratorCalm
About this appAI proposal
Hearthvoice turns scripts into speech and clones a voice from a short sample, and the models run on your own device in the browser. Your text and recordings are not sent to a server. There is no per-character meter, so a long audiobook costs the same as a short line.
It is made for video creators, podcasters, game makers and developers who generate a lot of speech and want to keep voice data private. Pick a voice, set the emotion for each line, render, and export audio files or a project you can reopen offline.
Unlike cloud services, usage is not billed by the character and the app keeps working without a connection once the voice models are downloaded. Unlike a bare open-source tool, it gives you a script editor, a voice library and batch export.
- Tagline
- Your voices, your machine, no meter running.
- Voice
- WarmPlainPrivate
- Why this name
- A hearth is a place you keep at home. The name says the voices and data stay with you, and the word voice says what it makes.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Runs on your device
Generation happens locally after a one-time model download. Text and voice samples stay on your machine.
No member yet
Clone from a short sample
Record or upload a clean clip, and get a reusable voice stored only in your browser storage or an export file.
No member yet
Emotion control per line
Mark each line as calm, excited, sad or stern, and adjust pace and intensity with simple sliders.
No member yet
No per-character fees
Generate as much as your hardware allows. Subscription or license pricing does not depend on text length.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#B4472BEmber · primary#FBF6EEParchment · surface#2A1F1ACharcoal · text#1F6F6BDeep Teal · accent#E0A43AAmber · data and highlights
Typography
Aa
Hearthvoice
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
Your voices, your machine, no meter running.
PricingAI proposal
The material gives no competitor prices, only that rivals charge per character. These tiers are a starting point to test, including whether a one-time license sells better than a subscription.
Free
Free
- Stock voices
- Short renders
- Watermark-free personal use
- No voice cloning
Creator
$12 a month
- Voice cloning
- Emotion control
- Batch render and export
- Commercial use license
Lifetime
$149 once
- Everything in Creator
- Updates for one year
- Commercial use license
- Voice pack updates
What it could be worthEstimate
What a software business making this much sells for today, from real deals. Pick a monthly revenue.
If it makes
Worth about
$194,000
Sale range $132,000 to $263,000
$120,000 a year in revenue
From 87 software businesses listed and sold on Empire Flippers in the last year, revenue vetted by the broker: they sell for about 19× their monthly revenue, most between 13× and 26×.
Measured on profit, the middle deal is 32× monthly net profit: about 2.7 years of profit.
What products in this market make
- ElevenLabsPublic claim$41,666,667 a month, worth about $808,966,000 at that multiple · cxodigitalpulse.com
An estimate, not an offer or financial advice. A real price depends on profit, growth, churn, how much of the work is the owner's, and the buyer.
How it would growAI proposal
How the product would reach new customers through its own use, and how the first ones would be found. A plan to test, not a result.
- Built-in loop
- Shared outputs. Creators publish audio made in the app to videos, podcasts and games. Exports can carry an optional 'Made with Hearthvoice' credit, and shared script links let a collaborator preview a line.
- Do they become customers?
- Listeners cannot tell how audio was made, so exposure is weak. Test whether the optional credit and preview links bring creators who then install the app and pay.
- First customers
- Solo YouTube and podcast creators who narrate in volume and already pay or feel the cost of per-character voice services.
- Where to reach them
- GitHub and open-source TTS communities: Post comparisons and demos where people already follow local TTS projects, and link to a free tier.
- Creator forums and subreddits: Share before and after workflows for narration in video and podcast communities, with a private-by-design angle.
- Indie game developer groups: Offer a batch dialogue render workflow for prototypes in game jam and dev communities.
- First milestone
- Twenty paying Creator or Lifetime customers from creator communities, with at least half rendering over an hour of audio in the first month.
Why people already pay for this
- VerifiedGitHub repository gained 908 stars in 26 days
- Public claimUsers avoid per-character fees that cloud APIs and proprietary software impose
- Public claimLocal processing without cloud dependency is valued
- Public claimVoice cloning and emotion control are features users want
- Public claim
Tech stack
Every proposal builds on the same PWA base, installable from the browser on any phone or computer. Dashed boxes are what this app adds.
- AppNext.js and TypeScript
- PWAManifest and service worker
- UITailwind CSS
- Data and authSupabase Postgres
- PaymentsStripe
- EmailResend
- HostingVercel
- This app · On-device inferenceWebGPU and WebAssembly speech model runtimeRuns TTS and cloning models inside the browser so no audio or text leaves the device.
- This app · Local storageIndexedDB and Origin Private File SystemCaches large model files and saved voices on the device for offline use.
RisksAI inference
- TechnicalBrowser memory and WebGPU limits may make good cloning slow or impossible on older phones and laptops. Model downloads may be several hundred MB and need clear device checks.
- CommercialFree open-source local TTS tools already exist. People may run them directly, so the editor, ease of setup and licensing must be worth paying for.
- Intellectual propertyOpen model licenses differ, and some bar commercial use or resale. Each bundled model needs a license check before launch.
- RegulatoryVoice cloning can be used to imitate people without consent. Consent prompts and terms are needed, and laws on synthetic voices vary by country.
- PlatformBrowsers may evict cached models or limit storage, especially on iOS, which could force repeated large downloads.
Everything marked AI proposal or AI inference was written by a model from the evidence on the opportunity page. The proposed name must be checked against trademark registers before it is used.