curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "text-to-speech-turbo-v2.5",
"text": "[whispers] I never knew it could be this way, but I'm glad we're here."
}
JSON
[sorrowful] I couldn't sleep that night. The air was too still, and the moonlight kept sliding through the blinds like it was trying to tell me something. [quietly] And suddenly, that's when I saw it.
curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "text-to-speech-turbo-v2.5",
"text": "[sorrowful] I couldn't sleep that night. The air was too still, and the moonlight kept sliding through the blinds like it was trying to tell me something. [quietly] And suddenly, that's when I saw it."
}
JSON
[tired] I've been working for 14 hours straight. [sigh] I can't even feel my hands anymore. [nervously] You sure this is going to work? [gulps] Okay… let's go.
curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "text-to-speech-turbo-v2.5",
"text": "[tired] I've been working for 14 hours straight. [sigh] I can't even feel my hands anymore. [nervously] You sure this is going to work? [gulps] Okay… let's go."
}
JSON
curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "text-to-speech-turbo-v2.5",
"text": "[laughing] Brooo—BROOO I don't know WHY that sent me!! [laughs harder] The chicken had NO PLOT, no twist, just raw determination!"
}
JSON
[frustrated sigh] I was seriously about to just trash the whole thing. And it was like... the FLOODGATES opened! It all just CLICKED. [happy gasp] I can't believe it actually worked.
curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "text-to-speech-turbo-v2.5",
"text": "[frustrated sigh] I was seriously about to just trash the whole thing. And it was like... the FLOODGATES opened! It all just CLICKED. [happy gasp] I can't believe it actually worked."
}
JSON
[professional] Thank you for calling Tech Solutions. How can I help you today? [sympathetic] Oh no, I'm really sorry to hear you're having trouble. [reassuring] Alright, based on what you're describing, I think we can fix this quickly.
curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "text-to-speech-turbo-v2.5",
"text": "[professional] Thank you for calling Tech Solutions. How can I help you today? [sympathetic] Oh no, I'm really sorry to hear you're having trouble. [reassuring] Alright, based on what you're describing, I think we can fix this quickly."
}
JSON
curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "text-to-speech-turbo-v2.5",
"text": "[strong French accent] Zat's life, my friend — you can't control everysing. [pause] But you can enjoy ze ride, non?"
}
JSON
Hold on, let me think. <break time="1.5s" /> Alright, I've got it. The key to understanding this phenomenon is not in the data itself, but in what the data leaves out.
curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "text-to-speech-multilingual-v2",
"text": "Hold on, let me think. <break time=\"1.5s\" /> Alright, I've got it. The key to understanding this phenomenon is not in the data itself, but in what the data leaves out."
}
JSON
curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "sound-effect-v2",
"text": "Distant thunder rumbling across a vast open plain during a summer storm, gradually building in intensity"
}
JSON
Speaker 1: [excitedly] Sam! Have you tried the new model yet? It's incredible! Speaker 2: [curiously] Just got access. What makes it so special? Speaker 1: [whispers] Watch this... [dramatically] I can do full Shakespeare now! Speaker 2: [with genuine belly laugh] Ha ha ha! That's amazing.
curl -X POST https://runapi.ai/api/v1/elevenlabs/text_to_sound \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "text-to-dialogue-v3",
"text": "Speaker 1: [excitedly] Sam! Have you tried the new model yet? It's incredible! Speaker 2: [curiously] Just got access. What makes it so special? Speaker 1: [whispers] Watch this... [dramatically] I can do full Shakespeare now! Speaker 2: [with genuine belly laugh] Ha ha ha! That's amazing."
}
JSON
FAQ
Working with Audio prompts
What makes a good %{model} prompt?
A useful %{model} prompt names the subject, style, constraints, and output intent clearly. The examples here are short enough to copy, but specific enough for an agent or backend job to preserve the generation goal.
Can I reuse these prompts across models?
Often, yes. Start with a prompt in this modality, then adjust model-specific fields such as aspect ratio, duration, voice settings, or style controls. The detail page shows any saved parameters next to the prompt text.
Where do I find the right model slug?
Every card shows the RunAPI model slug. Open the model page when you want only examples for one model, or follow the model catalog link for pricing and capability details before making a request.
Can agents call these prompts directly?
Yes. After installing the RunAPI MCP Server, paste the agent instruction from a prompt detail page. The page keeps the prompt text, model slug, and endpoint path together so the agent has enough context.