Synthesize speech directly from text.

post https://v1.api.audio/speech/sync

Good for time-critical applications. Maximum runtime is 30 seconds.

Caching

TTS responses are globally cached to improve performance. You can set Cache-Control to no-cache to skip the cache.
Following parameters are hashed as the cache key:

text
voice
speed
metadata
effect
bitrate
sampling_rate
output specified by the Accept header

Cache is missed when any of these parameters change.