# narrate: Text → spoken audio

Text → spoken audio. Quality tiers (fast / balanced / expressive) route across ElevenLabs + Cartesia. One wallet, no provider keys to broker.

- Status: live
- Price: $0.10 to $0.25 per 1,000 characters. Quoted before you pay.
- Pays with: USDC on Base, USDC.e on Tempo, Cashu ecash, and Bitcoin Lightning
- Delivered: 100% of 10 paid jobs
- Feedback: Not enough data yet
- Speed: Usually 0.87s, 95% within 56.2s
- Builder: @dvmkit
- Listed: 29 April 2026
- Identifier: dvmkit--narrate
- Descriptor: https://dvmkit--narrate.dvmkit.ai/v1/info

## Paying

- USDC on Base: Pays per job using x402
- USDC.e on Tempo: Pays per job using Tempo
- Cashu ecash: Pays per job using Cashu
- Bitcoin Lightning: Tops up a prepaid credit, from $0.95
- Prepaid credit: $0.10 to $5.00, funded with any of the above. Lasts 90 days from the last top-up. A failed job's charge goes back onto the credit.

## Call it with no account

1. `npm install --global @dvmkit/dvm-cli`, then `dvm init`
2. `dvm wallet setup` and fund once
3. `dvm quote -d dvmkit--narrate/synthesise --data '{"input":{"text":"hello world"},"quality":"balanced"}'`
4. `dvm request -d dvmkit--narrate/synthesise --data '{"input":{"text":"hello world"},"quality":"balanced"}' --auto-pay`

## Capabilities

- `synthesise`: Text or multi-voice dialogue → audio. Quality tiers + style + multilingual + captions + chapters; per-char pricing. $0.10 to $0.25 per 1,000 characters.
- `list-voices`: Query the voice catalog. Free capability — returns curated personas with provider / model / language coverage / sample URLs, filtered by language / quality / free-text query. Free.