❓┃help
No access to voices I got from the "List Voices" Endpoint
API
Bug Report
Can't play cartesia websocket data in web browser.
Feature Request
Bug Report
One word streaming broken
API
Playground
vqscore type
API
Hey, I'm having an issue getting a Studio narration exported.
Bug Report
Not reading now but was previously
Bug Report
Export narration from web after Pro upgrade never completes
Bug Report
Preview Voice
Feature Request
Websocket Connection Limit
API
Error creating key
Bug Report
NSFW
API
Hindi language: Unnecessary words are being spoken at the end of generated speech
API
Bug Report
Cartesia Player Stop function not working
API
Can't access a voice we love
Bug Report
Why there's weird end note in tts playback ??
API
Bug Report
broken audio on mobile phone device using sse endpoint.
API
Bug Report
trouble with delay while using cartesia.ai websocket
API
Using SSE with output format other than "raw" cause error.
API
Bug Report
Error using the python sdk (SSL: Certificate verify fail)
API
Sillytavern support?
Feature Request
API
Does the API support SSML tags like <speak> or <say-as>?
API
is `sonic-multilingual` available?
API
Region Pinning
Feature Request
Emotion sliders were removed from playground how to get same results in API
API
Playground
x-request-id -- is this a unique ID for each TTS task?
API
Cartesia in Livekit plugin throwing errors - how do I investigate?
API
SSE API - Understanding the response format
API
Why TTS request takes 6 seconds in my code while it takes <500ms to start playing in the playground?
API
Playground
Expected "embedding". Received "id".
API
Is there a way to check on a grant application?
Feature Request
How does requests work on cartesia pricing? Im on scale which includes 15 Parallel Requests
API
Where are the professional voice clone options? I see instant clone, but not the pro.
Playground
ETA on SSML support, and what will be supported?
Feature Request
Is creating a localized voice the only way to have a clone read out different languages?
Feature Request
Random pauses: even when there aren't any punctuations in the text
API
Bug Report
How to create voices? (E.g. school aged children)
Feature Request
Is there an asynchronous mode where the results are returned via webhook?
API
Unwanted sounds heard in the generated audio
API
Bug Report
API missing support for 'medium' emotions
API
Bug Report
High latency during audio generation using websocket
API
Bug Report
On Scale plan, 15 Parallel Requests - Does this mean only 15 Websocket con. can be open?
API
Conversion credit to TTS seconds
API
Surface "Preview URLs" of voices
Feature Request
Cannot Create API Key
API
Bug Report
Suppressing reading back Asterisk (*)
Bug Report
Websocket Idle CPU spikes
API
status 500 when voice cloning on interface
Bug Report
Playground
Volume inconsistency even when continuation is enabled
Bug Report
Latency issue (Node.js)
API
ModuleNot Found Error: Python API
API