AudioManager
class provides comprehensive audio functionality for MentraOS Apps, including URL-based audio playback, text-to-speech synthesis, and audio control. It automatically handles request tracking, timeouts, and provides real-time status monitoring.
options
: Audio playback configurationtext
: Text to convert to speech (required)options
: Text-to-speech configuration (optional)requestId
: Optional specific request ID to checkModel | Description | Languages | Latency |
---|---|---|---|
eleven_v3 | Human-like and expressive speech generation | 70+ languages | Standard |
eleven_flash_v2_5 | Ultra-fast model optimized for real-time use | All multilingual_v2 languages + hu, no, vi | ~75ms |
eleven_flash_v2 | Ultra-fast model (English only) | en | ~75ms |
eleven_turbo_v2_5 | High quality, low-latency with good balance | Same as flash_v2_5 | ~250-300ms |
eleven_turbo_v2 | High quality, low-latency (English only) | en | ~250-300ms |
eleven_multilingual_v2 | Most lifelike with rich emotional expression | en, ja, zh, de, hi, fr, ko, pt, it, es, id, nl, tr, fil, pl, sv, bg, ro, ar, cs, el, fi, hr, ms, sk, da, ta, uk, ru | Standard |
eleven_flash_v2_5
for real-time applications requiring ultra-low latencyeleven_turbo_v2_5
for a good balance of quality and speedeleven_multilingual_v2
for the highest quality emotional expressioneleven_v3
for maximum language support