Text to Speech

Turn 3-2,000 characters into a privately stored MP3 with ElevenLabs TTS Multilingual v2 through fal. Choose a stock voice and speed. The documented $0.10 per 1,000 characters becomes 1 credit per started 120-character block under Kanvora's existing formula.

Model
ElevenLabs TTS Multilingual v2fal · multilingual text to MP3 · stock voices · character-priced
Voice
Speed
Output
MP3 locked
Official audio proof reserved

Kanvora does not substitute another voice model's track as proof. This stage stays reserved until a source-tracked multilingual-v2 output is approved.

What is ElevenLabs text to speech?

Text to speech converts written text into one MP3 voice track. Kanvora exposes multilingual-v2, six stock voices, and three speeds only. It does not expose voice cloning, Music, eleven-v3, or text-to-dialogue. Noise Remover is a separate live private-upload tool.

Create a multilingual voice proof

  1. Prepare finished text

    Paste 3-2,000 characters and check names, numbers, punctuation, and abbreviations before generation.

  2. Choose a stock voice

    Select Rachel, Aria, Sarah, George, Charlotte, or Daniel. Kanvora does not accept custom or cloned voice IDs.

  3. Review the exact hold

    The quote updates from text length: 1 credit per started 120-character block, up to 17 credits at 2,000 characters.

  4. Listen before publishing

    Review pronunciation, pacing, tone, names, numbers, and usage rights in the privately stored MP3.

Tips for better results

  • Write punctuation for the pauses you want.
  • Spell out unusual names or numbers when pronunciation matters.
  • Use one stock voice and one speaking pace per request.
  • Review the entire MP3 before publishing or using it in another workflow.

Frequently asked questions

How is the credit cost calculated?

fal documents about $0.10 per 1,000 characters. Kanvora's $0.012 supplier-cost coverage per credit charges 1 credit per started 120 characters: 1,000 characters cost 9 credits and 2,000 cost 17.

Can I clone or upload a voice?

No. This route uses six stock ElevenLabs voices and rejects voice-cloning requests.

Does this include ElevenLabs music or sound effects?

Sound Effects v2 and Noise Remover are separate live Kanvora tools. Music, eleven-v3, speech-to-text, and text-to-dialogue are not shipped in this slice.

Which languages are supported?

The endpoint is ElevenLabs TTS Multilingual v2. Kanvora does not invent a complete language guarantee; test pronunciation in the stored proof.