ElevenLabs voice changer API: price, voices and code
The ElevenLabs voice changer takes a recording and gives back the same words, timing and emotion in another voice, for $0.003 a second. What a minute costs, the 54 voices, and the request that runs it.
結論
- Price: $0.003 a second of recording on Revana, as of November 2026: $0.03 for 10 seconds, $0.18 for a minute and $0.90 for five minutes.
- What it does: you send a recording and get the same words, timing and emotion back in another voice.
- Voices: 54 ElevenLabs voices, picked by id or by name.
- No subscription: prepaid credit from $10, and a job that fails is not charged.
ElevenLabs voice changer price
The voice changer is priced by the length of the recording you send, rounded up to the next whole second, at $0.003 a second. One request takes a recording of 1 to 300 seconds:
| Recording | 料金 |
|---|---|
| 10秒 | $0.03 |
| 30秒 | $0.09 |
| 1分 | $0.18 |
| 5 minutes | $0.90 |
For a recording longer than 5 minutes, split it into parts and send each part as its own request. Every other model's price is on the pricing page.
ElevenLabs text to speech prices
To make speech from text instead of a recording, use one of the ElevenLabs text to speech models. They are priced per 1,000 characters, with up to 5,000 characters a request:
| モデル | モデルID | Per 1,000 characters |
|---|---|---|
| ElevenLabs V4 | elevenlabs-v4 | $0.21 |
| ElevenLabs V3 | elevenlabs-v3 | $0.21 |
| ElevenLabs Multilingual V2 | elevenlabs-multilingual-v2 | $0.21 |
| ElevenLabs Turbo 2.5 | elevenlabs-turbo-2.5 | $0.21 |
More on these models in ElevenLabs API pricing.
Which voices you can use
Send the voice in voice, as a voice id or by name, such as Brian. Names are matched without regard to capitals; where two voices share a name, send the id. Leave voice out and James reads the recording. Every voice id, with a short description, is in GET /v1/models under options.voices. A few of the 54:
- James: Husky, Engaging and Bold
- Brian: Deep, Resonant and Comforting
- Bella: Professional, Bright, Warm
- Callum: Husky Trickster
- Laura: Enthusiast, Quirky Attitude
- Nathaniel: Engaging, British and Calm
- Allison: Calm, Soothing and Meditative
- Hank: Deep and Engaging Narrator
You can also send stability and similarity_boost, each from 0 to 1, to tune the voice. Only send recordings you have the right to use.
Call the ElevenLabs voice changer on Revana
Create a key on the sign-up page and send the model id elevenlabs-voice-changer to POST /v1/generations. The request format is in the docs, every option and its price on the ElevenLabs Voice Changer page, and every model and price on the models page and the pricing page.
Put the recording in audio_url as a public https link; if the file is on your computer, upload it first with POST /v1/uploads. This request turns a one-minute recording into James's voice, which costs $0.18, and the finished job links to an MP3:
export REVANA_API_KEY="sr_live_..."
# One Idempotency-Key per job, made once: resend with the same $KEY after a timeout, a 429 or a 5xx
KEY=$(uuidgen)
# 1. Submit the job (returns 202 with the generation id). max_cost_usd is your cap: the most it may cost.
curl https://api.revana.si/v1/generations \
-H "Authorization: Bearer $REVANA_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $KEY" \
-d '{
"model": "elevenlabs-voice-changer",
"input": {
"audio_url": "https://example.com/my-take.mp3",
"voice": "EkK5I93UQWFDigLMpZcX"
},
"max_cost_usd": 0.2
}'
# 2. Poll with the id from step 1 until status is "succeeded" or "failed".
# A failed job is free; make a new KEY before you submit it again.
curl https://api.revana.si/v1/generations/gen_01JC4K9R2T \
-H "Authorization: Bearer $REVANA_API_KEY"よくある質問
How much does the ElevenLabs voice changer API cost?
On Revana, $0.003 a second of recording, rounded up to the next whole second, as of November 2026. A one-minute recording costs $0.18 and a five-minute one $0.90.
How long can the recording be?
From 1 to 300 seconds in one request. Split a longer recording into parts and send each one as its own request.
Which voices can I use?
Any of the 54 ElevenLabs voices listed in GET /v1/models under options.voices, by id or by name. James is the default.
Do I need an ElevenLabs subscription?
No. You call the voice changer with a Revana API key and pay per second from prepaid credit, with no monthly plan.
Am I charged if the voice change fails?
いいえ。料金は送信時に確保され、ジョブが失敗した場合は解放されます。