Skip to content
voice-goio

voice-goio docs

Voice Assignment

The speaker does this itself. An application that assigns voices on its own uses a voices.Pool:

pool := voices.NewPool(man, voices.PoolOptions{Dir: voices.Dir(), FemaleShare: 1.0 / 9})
twr := pool.Assign("LKPR", voicegoio.Tower)          // the same voice all session
gnd := pool.Assign("LKPR", speaker.KindOf("ground")) // never the tower's speaker
crew := pool.AssignCrew("CSA123")                    // never a controller's voice

The rules

  • A position keeps its voice. A controller position at an airport keeps its voice, and two positions at the same airport never share one. The speaker hands a position over to a new voice every 30 to 60 minutes (a shift change).
  • One controller, one voice on every frequency (v0.13.0). Setting Utterance.Controller keys the voice by the person rather than the position: ground and tower worked by one controller sound like one person, with the radio sound of the Position each call is made on.
  • Crews never get a controller’s voice (v0.13.1). Pool.AssignCrew keeps a crew’s voice for the session, with a cockpit’s radio sound, and keeps controller and crew voices apart while either side has a free one. The speaker uses it for every pilot call.
  • Regional. Assignment is region-weighted; Assign records the accent it chose on VoiceProfile.Accent. Controller accents in the speaker are opt-in (speaker.Options.Accents, off by default).

The ATIS

The ATIS gets one fixed machine voice. It is not assigned from the regional pool and does not consume an airport’s voices: an ATIS is a recording played on a loop, so it is the same flat voice everywhere, and giving Gatwick a different ATIS reader from Heathrow would be wrong rather than varied.

Unaudited speakers

Speaker-level audit (quality, pass in the manifest) is a human listening pass, so the manifest ships those fields unset rather than fabricated. Until a model is audited its speakers are only assignable with PoolOptions.AllowUnaudited. voicecheck audit -model <name> renders the audit text in every speaker of a model for that pass.