speech to text agents

6 tagged speech to text, measured the same way as everything else here.

Browse within: Apple Silicon 6asr 6bun 6coreml 6openclaw 6

audio-quality-check

01

drakulavich/kesha-voice-kit

Agent Claude Code

Use after any commit touching rust/src/tts/ to objectively check that the TTS engine still produces sane audio for a fixed Russian + English corpus. Computes RMS, silence ratio, sample rate, channel count, length-vs-text ratio. Flags suspicious WAVs (all-silence, clipping, wrong rate, monosamples, length 10x off vs…

73 2d ago A 105 tokens original MIT

implementer

02

drakulavich/kesha-voice-kit

Agent Claude Code

Take one ticket from plan to draft PR — plan first and stop for the team lead's verdict, then implement the approved plan test-first in a worktree. Never implements an unapproved plan.

73 2d ago A 42 tokens original MIT

teamlead

03

drakulavich/kesha-voice-kit

Agent Claude Code

Own a ticket end to end — size it, spawn the implementer, judge its plan, run the review round, triage the findings, hand the pull request over, and keep the ledger. Never authors what it judges.

73 2d ago A 49 tokens original MIT