Murf API
Budget application speech synthesis by characters and separately priced API endpoints.
- Starting price
- Falcon USD 0.01/1,000 chars
- Free access
- USD 10 free API trial credit
- Billing unit
- Characters; endpoint-specific rates
- Platform
- HTTP API integration
- Verified
- 20 Oct 2026
Overview
What you provide
- Provide text, a supported voice selection and API credentials from your own account.
What you get
- Murf API converts application text into generated speech with separately metered synthesis and translation endpoints.
Choosing this tool
Murf API is useful for adding speech generation to an application with explicit character metering. This page evaluates the API rather than the Studio editor. Piper is an alternative local synthesizer when you can host voice models; Kokoro offers another local inference route with a different integration burden.
A decision worked through
At the published Falcon rate, 100,000 characters cost USD 1; at Gen2 rates they cost USD 3, before any other endpoint usage. Translation of the same character count is listed separately at USD 2, so a translate-then-speak workflow needs both line items.
Murf API alternatives for this task
For Murf API, compare Piper, Kokoro-82M. Match the input, delivery format and operating responsibilities before comparing price.
| Product | Price / unit | Free allowance | Input → output | Key limit |
|---|---|---|---|---|
| Murf APIRead profileOfficial source ↗ | Falcon USD 0.01/1,000 chars | USD 10 API credit | Application text becomes audio. | Studio plans are separate. |
| PiperOfficial source ↗ | Local open-source runtime | Use your own installed voice models | Text becomes locally synthesized speech. | Host runtime and review voice licenses. |
| Kokoro-82MOfficial source ↗ | Apache-2.0 weights; own compute | Local weights, no hosted quota promised | Text becomes model-generated audio. | Requires an inference integration. |
Setup & workflow
- Create an API trial and choose the endpoint and voice required by the app.
- Send a short authorized text sample and review the returned speech.
- Track characters and concurrency separately before scaling the integration.
Requirements & installation
- An application must integrate the HTTP API and store its own key securely.
Limits & review
- Five concurrent requests are listed for trial and pay-as-you-go. Character limits and voice entitlements depend on the endpoint; no production latency benchmark was run for this profile.
Your part
- Check pronunciation, latency and authorization for the selected voice before deployment.
Sources & verification5
This profile is based on official sources, not a hands-on product test.
Vendor descriptions and demos document advertised features. Editorial guidance is based on these sources.
- Official API endpoint ratesmurf.ai
- Official USD billing currencyhelp.murf.ai
- Murf API official referencemurf.ai
- Piper official referencegithub.com
- Kokoro-82M official referencehuggingface.co
