// verdict #13 · 23 May 2026 · 14:08 UTC

github.com/supertone-inc/supertonic

HOLD

the voices disagreed — or rivera asked for redesign before adoption. held for more information, or wider reach.

submitted by Rob · BLKOUT · model anthropic/claude-sonnet-4 · against wishlist of 2026-04-27T11:18:37z · visit the tool → · back to history

// the four voices, in parallel

What the Council said

// 01 · critic

Baldwin

hold

technical risk, lock-in, sequencing — the master's tools.

Looking at Supertonic, this is a text-to-speech system that runs locally via ONNX Runtime. Let me stress-test this against BLKOUT's context and guardrails. **GUARDRAIL VIOLATIONS:** None directly. This doesn't touch the CRM schema, doesn't require n8n, and doesn't assume always-on e-commerce. However, it does add operational complexity to infrastructure that needs stability first. **VENDOR LOCK-IN ASSESSMENT:** Actually strong here. Open-weight model (99M parameters), runs via ONNX Runtime, multiple SDK options, and can be self-hosted. The models are on Hugging Face with clear licensing. No dangerous dependencies on Supertone's commercial services — the core inference runs locally. **TECHNICAL DEBT CONCERNS:** This is where it gets problematic. BLKOUT's Tier 1 priority is "tech stability and system resilience" — the VPS, Coolify, and all modules need to be reliable before adding more capabilities. A TTS system adds: - New Docker containers or processes to manage - Audio file storage and serving - Additional memory/CPU overhead (99M parameters still requires resources) - Another failure point in the stack **SEQUENCING PROBLEMS:** The wishlist mentions "AI-generated audio (NotebookLM Deep Dives)" under the podcast hosting addition, but TTS isn't explicitly called out as a current need. This feels like solution-first thinking rather than addressing an articulated problem. **OPERATIONAL OVERHEAD:** For a small team, this adds complexity without clear immediate value. Audio generation, storage, cleanup, and integration with existing systems all require ongoing maintenance. The "lightning-fast" claims don't address the operational cost of keeping it running reliably. **WISHLIST MAPPING:** Doesn't clearly map to any of the fifteen items. Closest might be supporting future podcast content, but that's speculative. The podcast hosting addition mentions NotebookLM audio, which suggests the AI audio generation need is already being met elsewhere. **SURVEILLANCE MECHANICS:** Clean on this front — runs locally, no cloud calls, no data extraction. **THE REAL QUESTION:** What problem is this solving that BLKOUT actually has right now? The infrastructure isn't stable enough to support nice-to-haves, and there's no clear community need being articulated for locally-generated TTS. This is a well-engineered tool that doesn't violate guardrails, but it's the wrong priority at the wrong time for an organization that needs to focus on foundational stability before adding capabilities. RECOMMENDATION: HOLD

// 02 · ethicist

Murray

pass

values, sovereignty, community benefit.

I need to evaluate Supertonic — a text-to-speech system — against BLKOUT's values and strategic priorities. **What Supertonic Is:** A local, on-device text-to-speech system that converts text to audio in 31 languages. It runs entirely offline via ONNX Runtime, requires no cloud services, and offers voice cloning capabilities through their "Voice Builder" service. **Values Alignment Assessment:** **DATA SOVEREIGNTY: STRONG** — The core system runs entirely on-device with no cloud dependency. Text processing and audio generation happen locally, meaning BLKOUT retains complete control over any content processed. **ANTI-SURVEILLANCE: MIXED** — The base system is surveillance-free since it's offline. However, the "Voice Builder" service for custom voices requires uploading voice samples to Supertone's servers, which introduces a data sovereignty concern for any voice cloning use cases. **COMMUNITY BENEFIT: NEUTRAL** — TTS could serve accessibility needs, but this isn't obviously aligned with BLKOUT's specific mission to Black queer men and their communities. **OPEN-SOURCE ALIGNMENT: PARTIAL** — The model weights are open, but it's not fully open-source. The ONNX runtime dependency is open, but Supertone Inc. controls the model development. **Wishlist Mapping:** This doesn't clearly map to any specific wishlist item. The closest connection might be to **"podcast hosting"** — the new wishlist item about publishing AI-generated audio content. TTS could generate audio from text content, but BLKOUT already has NotebookLM for Deep Dives, and this would be adding complexity rather than solving the stated need for hosted distribution and analytics. **Guardrails Check:** - **Schema ownership**: Not applicable — this doesn't touch member data - **Pre-order model**: Not applicable - **Infrastructure before application**: This would add operational complexity without clear infrastructure benefit **Strategic Concerns:** 1. **Mission alignment unclear** — TTS technology doesn't obviously serve Black queer community needs 2. **Solution looking for problem** — No clear use case articulated in current BLKOUT priorities 3. **Capacity distraction** — Would require learning new technology stack without clear community benefit 4. **Voice ethics** — Voice cloning technology raises questions about consent and authenticity that would require the **"AI policy"** wishlist item to be completed first **The fundamental question:** What community need would this serve that isn't already met or planned? The podcast hosting need is about distribution infrastructure, not content generation. Member communication needs are met through existing channels. This appears to be interesting technology without a clear application to BLKOUT's mission or current priorities. It would add complexity without addressing any identified community need or wishlist item. RECOMMENDATION: PASS

// 03 · builder

Rustin

vibe-codeability, stack fit, lift — what'll actually ship.

Looking at Supertonic, this is a text-to-speech system that runs locally via ONNX Runtime. Let me break down the buildability assessment: ## VIBE-CODEABILITY **Strong.** The documentation looks comprehensive with multi-language SDK support. Python integration via `pip install supertonic` is straightforward. The fact they have working examples across Python, Node.js, and browser implementations suggests good DX. ONNX Runtime is well-established with solid docs. ## STACK COMPATIBILITY **Excellent.** ONNX Runtime containerizes cleanly and runs on CPU (no GPU requirement). This should deploy smoothly on Coolify with minimal resource overhead. The 99M parameter model is compact enough for VPS deployment. ## SPEED OF DEPLOYMENT **Fast.** Python SDK install, download pre-trained models from Hugging Face, and you're generating audio. The "local HTTP server with `/v1/tts` endpoints" feature means it can integrate as a microservice immediately. ## LIFT ASSESSMENT **Accessible.** This maps to the **podcast hosting** addition to the wishlist. While the wishlist asks for hosted podcast platforms, Supertonic could enable BLKOUT to generate audio content locally (like the NotebookLM Deep Dives mentioned) without dependency on external TTS services. The integration work is straightforward Python scripting. ## MAINTAINABILITY **Good.** Once deployed, it's essentially a local API server. Failure modes are predictable (model files, disk space, memory). Recovery doesn't require specialist knowledge. Updates come via model downloads, not complex migrations. ## INTEGRATION POTENTIAL **High.** Could integrate with: - Commons pages (generate audio versions of written content) - CRM (accessibility features for member communications) - AIvor (voice responses to member needs) - Events (audio announcements, accessibility support) The `/v1/audio/speech` OpenAI-compatible endpoint means it slots into existing workflows easily. ## GUARDRAIL CHECK **Clean.** No guardrail violations. Doesn't duplicate existing functionality, doesn't require n8n, maintains data sovereignty (runs locally), and fits the infrastructure-first approach. ## STRATEGIC ALIGNMENT This doesn't directly solve podcast hosting, but it enables **content accessibility** and **audio-first member experience** - both aligned with BLKOUT's community benefit mission. The 31-language support could serve diverse community needs. Local deployment means no privacy concerns with member content. The tool is production-ready, well-maintained (active updates through 2026), and the company offers both open-source and commercial tiers without lock-in. **RECOMMENDATION: GO**

// 04 · inclusion

Rivera

reach, the under-served, who's on the street.

Looking at Supertonic through the lens of who gets left behind and who gets brought in. **WHO IS THIS TOOL NOT FOR?** **ACCESSIBILITY FLOOR**: This tool actually lowers several barriers. It runs locally without internet dependency, works on mobile devices and even Raspberry Pi, and supports 31 languages including Arabic, Hindi, and others often excluded from tech tools. The "language-agnostic" mode (`lang="na"`) means someone doesn't need to know or declare what language they're using. That's inclusion-first design. **PRIVACY ARCHITECTURE**: This is where Supertonic shines for BLKOUT's context. Completely local processing means no data leaves the device — critical for closeted members, asylum seekers, or anyone in unsafe situations who needs text-to-speech without creating a digital trail. No cloud dependency, no API calls, no privacy concerns. **DEVICE AND COST ASSUMPTIONS**: Runs on phones, browsers, and low-resource hardware. The 99M parameter model is specifically designed to be lightweight. No subscription fees, no API costs after initial setup. This doesn't assume expensive hardware or ongoing payments. **SYNC vs ASYNC**: Pure async tool — generate audio when you need it, no live presence required. Respects variable capacity and different schedules. **REGISTER AND LANGUAGE**: 31 languages is substantial reach beyond English. The language-agnostic processing means it doesn't assume linguistic confidence or formal education about language identification. **MAPPING TO BLKOUT WISHLIST**: This directly serves the **podcast hosting** need that was just added to the wishlist in April 2026. BLKOUT is starting to publish AI-generated audio and needs better distribution than Drive embeds. Supertonic could generate accessible audio versions of written content, create multilingual versions for diaspora reach, or enable members to create audio content without needing recording equipment or studio access. **GEOGRAPHIC REACH**: The multilingual support and offline operation make this work anywhere — Manchester, Birmingham, Cardiff, Glasgow, rural areas, and internationally. No London-centric assumptions. **POTENTIAL CONCERNS**: The voice cloning capability raises questions about consent and impersonation. The "Voice Builder" feature could be misused. However, the core TTS functionality doesn't require cloning — it has preset voices for basic text-to-speech needs. **ISOLATION vs NETWORK**: This helps isolated individuals create audio content without needing recording skills, studio access, or technical confidence. Someone can write text and generate professional-quality audio entirely privately on their device. This tool genuinely extends reach to under-served communities: people without reliable internet, those needing privacy protection, speakers of non-English languages, people without recording equipment or technical audio skills, and those in geographic areas with poor connectivity. The privacy-first, multilingual, low-resource design pattern is exactly what inclusion looks like in practice. RECOMMENDATION: GO

// context at time of judgement

The Wishlist, as it was then +

this is the document the council was asked to hold this tool against. click to expand.

BLKOUT Digital Strategy — Living Document

This document is the authoritative context for the Dreamcatcher Council. It combines BLKOUT's Year One guardrails with the fifteen-item wish list. The Council reads this fresh on every evaluation — edits here immediately reshape every future verdict.

Last revised April 2026. For board and community input.

Year One Guardrails

Rules that hold while foundational infrastructure is being built. Harder than tier priorities — these do not negotiate.

— No n8n. All automations use node-cron or Supabase triggers. n8n was tried and removed; the operational cost of a workflow engine outweighed the benefit for a small team.

— Schema ownership. The CRM's 001_crm_complete_schema.sql is authoritative. Any new tool that touches member data reads from CRM tables — it does not maintain a separate schema.

— Pre-order model, not always-on store. The BLKOUT shop is a waiting-list / pre-order surface with CBS tiers (free / £3 per month / £10 per month). Tools that assume always-on e-commerce are a category mismatch.

— Stripe pre-flight. Stripe routes stay disabled until credentials are confirmed in Coolify. Test mode only — no tool that requires live payment provisioning on adoption.

— CRM mock data stays until migrations verified. Nothing removes CRM mock data until the equivalent migration has been confirmed via supabase-query.mjs. Tools that auto-migrate are held until verification is possible.

— RSVP and Stripe are already built. Frontend and backend exist in the platform. Tools that duplicate these are a category mismatch — the gap is credentials and tables, not code.

— Infrastructure before application. Debug infrastructure (CPU, disk, running services, Docker images) before application code. Tools that add operational complexity must justify themselves against this order.

Wish List — fifteen capabilities

Prioritised by potential impact and capacity to instigate change. First draft March 2026, living document since.

Tiers: do now (accessible) — plan and build (moderate lift) — develop with intent (significant undertaking)

Categories: infrastructure — member experience — co-production

Tier 1 — do now

High impact, low barrier — no dependency blockers.

— AI policy, à la New Public infrastructure accessible Unlocks the integrity of everything beneath it. Determines what BLKOUT will and won't do with AI, and why. Data sovereignty and privacy follow from this document, not the other way around. Mainly a thinking and writing exercise — the hardest part is the thinking.

— CRM — values-embedded member system member experience accessible Already in active use. Develop with transparency and member control principles baked in before habits calcify around the wrong defaults. Needs connecting to member pages, AIvor, and the directory as those mature.

— Leania.ai — AI ops audit infrastructure accessible Quick diagnostic of workflows, tools and bottlenecks. Keep / kill / replace / automate recommendations. Low cost, high information value. Caveat: recommendations will skew commercial — use the diagnostic layer, not the prescription.

— Tech stability and system resilience infrastructure moderate lift Without this, everything built on the infrastructure will fail. The VPS, Coolify, and all modules need to be reliable and recoverable before more is added. Foundational before ambitious.

Tier 2 — plan and build

High impact, requires resourcing, people or sequencing.

— Bookkeeping, financial regulations and fundraising infrastructure moderate lift CBS obligation, funder pipeline active. Needs the right person as much as the right tool — the Treasurer role is load-bearing. Required before significant grant income flows through the organisation.

— Compliance infrastructure — demonstration not monitoring infrastructure moderate lift Mission risk if left too long. Evidence of governance, financial stewardship, and community benefit must be buildable in real time, not reconstructed after challenge. Systems need building before they are needed.

— Regular events calendar and Brunch 2030 member experience moderate lift Visible, reliable rhythm signals organisational health. Monthly social, film screening at The Arzner, community organisers gathering, seasonal specials, Berto Pasuka day. Key venue relationships held — Stanley Arts, The Arzner. Delegate the programming.

— Non-intrusive lead generation member experience moderate lift Community-appropriate discovery. Transparent about the win on all sides. Values-consistent methodology — games welcome if honest, surveillance mechanics excluded. The question is whether the constraint is capacity, visibility, or trust, because each has a different solution.

— Live directory of Black queer civil society co-production moderate lift Co-production and community research — built by the community, kept live by the community. Feeds CRM, AIvor, lead generation, and network mapping simultaneously. The gaps in the directory are as revealing as what is in it.

— Partner tool and project management dashboard infrastructure moderate lift Operational capacity for transparent relationships with organisations of all sizes. Sits on stable infrastructure. Enables partnerships to be held without depending on founder memory.

— Network mapping — loneliness co-production moderate lift How loneliness manifests for Black queer men — which responses are proximate, which require greater effort. Research question still needs articulating before design. Board community lead to own this. Brunch network exercise as potential methodology.

Tier 3 — develop with intent

Transformative — significant undertaking, dependencies to resolve first.

— Unique member pages — member-controlled data as feature member experience significant Members see exactly what BLKOUT holds about them, can edit or delete any of it, and are equipped to ask the same of others. Policy as education. Significant data architecture. Needs CRM foundation first. The feature that most visibly demonstrates the values.

— Next-level AIvor — orchestration on member need member experience significant From broadcaster to responder — receiving member need signals and acting on them, with care and ethical guardrails. The line between care and surveillance must be designed in from the start. Depends on knowing what members actually need. Network mapping and directory must precede.

— Co-designed member consultation with iterative learning loop co-production significant Design the consultation process in consultation. Radical transparency, shared results, member agency in decisions that follow insight. Includes autoresearch-style closed feedback cycle with co-designed metrics. Never a failure, always a lesson — at organisational scale. The most ambitious item on the list and the one that could make every other item more effective.

— Youth leadership intervention co-production significant Form and content both need co-producing — this cannot be designed for young people, only with them. Requires funding. Can begin relationship-building and scoping conversations now, against the backdrop of Black youth unemployment returning to 1981 levels.

— Enterprise and wealth — pop-up / shebeen / art model co-production significant Still forming. The relationship between enterprise, workplaces, community and intergenerational wealth remains to be discovered. An inquiry that in its form embodies potential elements of the final proposal — collaborative, cultural, accessible, profitable, sociable. Trust for London proposal as the discovery vehicle.

BLKOUT Creative Ltd — Community Benefit Society Living document — edited in situ via the Dreamcatcher editor

---Podcast hosting — added April 2026, surfaced via Commons page build

What we want: A hosted podcast platform with a brand-themeable embeddable player (we'll skin to obsidian/gold), an RSS feed for distribution to Spotify/Apple Podcasts, and basic listener analytics. Should integrate with the Commons + Critical Frequency pages without a generic Drive iframe.

Why now: We're starting to publish AI-generated audio (NotebookLM Deep Dives, future episodes from the Compass material). Drive embed is the current workaround — generic player, no podcast-app distribution, no analytics, asset lives in personal Drive.

Constraints / what we won't: Free tier or low monthly cost (BLKOUT is small); no extractive licensing on community audio; option to self-host or export RSS to migrate (no lock-in); accessible player UI; no Spotify exclusivity.

On the radar for Council review when surfaced: Castopod (open source, self-hostable on Coolify — natural fit), Transistor (paid, neutral, RSS-first), Buzzsprout, Spotify for Podcasters / Anchor (free + reach but sovereignty concerns).