lumen-ai/com.subhashh.lumen.ai/SpeechModel

SpeechModel

Subtitles: Cactus Compute's Whistle (speech to text, Apache 2.0) and Silero VAD v6.2.3 (where people speak, MIT), so music isn't "transcribed" into invented lines. Unchanged copies, hosted together; downloaded with either Smart Search tier.

Types

Tier

class Tier(val id: String, val label: String)

Settings' choices, by quality: Whistle for now; Whisper or Parakeet if they earn a place.

Properties

bytes

val bytes: Long

Commit

const val Commit: String

files

val files: List<ModelFile>

Repo

const val Repo: String

tiers

val tiers: List<SpeechModel.Tier>

Functions

dir

fun dir(context: Context): File

installed

fun installed(context: Context): Boolean

url

fun url(file: ModelFile): String