Bland Speech
Bland's text-to-speech model, built for phone calls and kept natural with breaths, stumbles and pauses, used in a web studio or through one streaming API endpoint, with voice cloning and a director that writes and performs scripts from a scene description, billed per character.
Find alternatives to Bland Speech
What it does
- Text to speech
Features 4
- Voice cloningInstant clone from about 10 seconds of audio, or a professional clone fine-tuned on 30+ minutes of audio.
- Performance tagsBracketed instructions such as [laughs] or [clears throat] that the model performs instead of reading aloud.
- Speech studio with directorA studio to try voices in, where a director writes the dialogue, stumbles and timing from a scene description.
- Streaming text-to-speech APIA /v1/speak endpoint that streams audio back over HTTP chunks or a WebSocket.
Sources
- Pricing
-
Official documentation · 10 Oct 2026
"$0.04 per 1,000 characters. Pay as you go." labelled a "Limited-time launch offer"; new accounts get 50k characters free once; loading $5 unlocks professional voice cloning and 5x concurrency. A one-off starting allowance, not a recurring free tier.
- Description
-
Official documentation · 10 Oct 2026
Heading "The most human text-to-speech model." FAQ: "Bland Speech turns text into speech that sounds like a person." Studio for trying it and an API endpoint, POST /v1/speak, streaming over HTTP chunks or a WebSocket; professional voice cloning; a "director" that writes and performs scripts from a scene description.
- What it does
-
Official documentation · 10 Oct 2026
Turns text into spoken audio.
- Sold within
-
Official documentation · 10 Oct 2026
Its own pay-as-you-go rate on its own page, the same in the studio and the API, with self-serve sign-up — a standalone route.
- Name
-
Official documentation · 10 Oct 2026
Page title "Bland Speech | Bland AI"; the current version is called Bland Speech v3.
- URL
-
Official documentation · 10 Oct 2026
The product's own page, linked from the homepage navigation.
- Deployment
-
Official documentation · 10 Oct 2026
Hosted studio and API.
- Interfaces
-
Official documentation · 10 Oct 2026
Web studio and REST/WebSocket API.
- Intended for
-
Developers building with the endpoint and teams producing voice audio in the studio.
Also from Bland AI
-
Bland Agent platform
Voice AI platform for building, testing and running AI phone agents that handle inbound and outbound calls, with the same agent able to work over SMS, iMessage and web chat, an API, conversational pathways, and per-minute pricing that includes the language model, speech-to-text and text-to-speech.
-
Norm Agent platform
An AI assistant inside the Bland platform that builds a production voice agent from a plain-English description, generating its conversation pathway, persona, voice and tools, with templates for support, appointment booking, lead qualification and outbound sales calls.
Open Bland AI in the directory
Something wrong here? Send a correction — quote this product id: bland-speech.