Documentation

TTS extends Message
in package

The agent's voice.

Generated from protobuf message livekit.SimulationRun.JobMetrics.TTS

Table of Contents

Properties

$enunciation_score  : mixed
judged 0-1: key entities audibly intact
$naturalness_score  : mixed
judged 0-1: prosody / expressiveness
$speech_rate_wpm  : mixed
speaking rate; conversational English ~110-150
$ttfa_ms  : mixed
Time to first AUDIO — the user-relevant signal. Some providers stream bytes before actual speech, so byte-level TTFB alone over-promises; it is kept below for disambiguation.
$ttfb_ms  : mixed
provider byte-level TTFB, agent-reported
$wer  : mixed
intelligibility: intended text vs a transcription of the output audio

Methods

__construct()  : mixed
Constructor.
clearEnunciationScore()  : mixed
clearNaturalnessScore()  : mixed
clearSpeechRateWpm()  : mixed
clearTtfaMs()  : mixed
clearTtfbMs()  : mixed
clearWer()  : mixed
getEnunciationScore()  : float
judged 0-1: key entities audibly intact
getNaturalnessScore()  : float
judged 0-1: prosody / expressiveness
getSpeechRateWpm()  : float
speaking rate; conversational English ~110-150
getTtfaMs()  : int
Time to first AUDIO — the user-relevant signal. Some providers stream bytes before actual speech, so byte-level TTFB alone over-promises; it is kept below for disambiguation.
getTtfbMs()  : int
provider byte-level TTFB, agent-reported
getWer()  : float
intelligibility: intended text vs a transcription of the output audio
hasEnunciationScore()  : mixed
hasNaturalnessScore()  : mixed
hasSpeechRateWpm()  : mixed
hasTtfaMs()  : mixed
hasTtfbMs()  : mixed
hasWer()  : mixed
setEnunciationScore()  : $this
judged 0-1: key entities audibly intact
setNaturalnessScore()  : $this
judged 0-1: prosody / expressiveness
setSpeechRateWpm()  : $this
speaking rate; conversational English ~110-150
setTtfaMs()  : $this
Time to first AUDIO — the user-relevant signal. Some providers stream bytes before actual speech, so byte-level TTFB alone over-promises; it is kept below for disambiguation.
setTtfbMs()  : $this
provider byte-level TTFB, agent-reported
setWer()  : $this
intelligibility: intended text vs a transcription of the output audio

Properties

$enunciation_score

judged 0-1: key entities audibly intact

protected mixed $enunciation_score = null

Generated from protobuf field optional float enunciation_score = 6;

$naturalness_score

judged 0-1: prosody / expressiveness

protected mixed $naturalness_score = null

Generated from protobuf field optional float naturalness_score = 5;

$speech_rate_wpm

speaking rate; conversational English ~110-150

protected mixed $speech_rate_wpm = null

Generated from protobuf field optional float speech_rate_wpm = 4;

$ttfa_ms

Time to first AUDIO — the user-relevant signal. Some providers stream bytes before actual speech, so byte-level TTFB alone over-promises; it is kept below for disambiguation.

protected mixed $ttfa_ms = null

Generated from protobuf field optional uint32 ttfa_ms = 1;

$ttfb_ms

provider byte-level TTFB, agent-reported

protected mixed $ttfb_ms = null

Generated from protobuf field optional uint32 ttfb_ms = 2;

$wer

intelligibility: intended text vs a transcription of the output audio

protected mixed $wer = null

Generated from protobuf field optional float wer = 3;

Methods

__construct()

Constructor.

public __construct([array<string|int, mixed> $data = null ]) : mixed
Parameters
$data : array<string|int, mixed> = null

{ Optional. Data for populating the Message object.

@type int $ttfa_ms
      Time to first AUDIO — the user-relevant signal. Some providers stream
      bytes before actual speech, so byte-level TTFB alone over-promises;
      it is kept below for disambiguation.
@type int $ttfb_ms
      provider byte-level TTFB, agent-reported
@type float $wer
      intelligibility: intended text vs a transcription of the output audio
@type float $speech_rate_wpm
      speaking rate; conversational English ~110-150
@type float $naturalness_score
      judged 0-1: prosody / expressiveness
@type float $enunciation_score
      judged 0-1: key entities audibly intact

}

clearEnunciationScore()

public clearEnunciationScore() : mixed

clearNaturalnessScore()

public clearNaturalnessScore() : mixed

clearSpeechRateWpm()

public clearSpeechRateWpm() : mixed

clearTtfaMs()

public clearTtfaMs() : mixed

clearTtfbMs()

public clearTtfbMs() : mixed

clearWer()

public clearWer() : mixed

getEnunciationScore()

judged 0-1: key entities audibly intact

public getEnunciationScore() : float

Generated from protobuf field optional float enunciation_score = 6;

Return values
float

getNaturalnessScore()

judged 0-1: prosody / expressiveness

public getNaturalnessScore() : float

Generated from protobuf field optional float naturalness_score = 5;

Return values
float

getSpeechRateWpm()

speaking rate; conversational English ~110-150

public getSpeechRateWpm() : float

Generated from protobuf field optional float speech_rate_wpm = 4;

Return values
float

getTtfaMs()

Time to first AUDIO — the user-relevant signal. Some providers stream bytes before actual speech, so byte-level TTFB alone over-promises; it is kept below for disambiguation.

public getTtfaMs() : int

Generated from protobuf field optional uint32 ttfa_ms = 1;

Return values
int

getTtfbMs()

provider byte-level TTFB, agent-reported

public getTtfbMs() : int

Generated from protobuf field optional uint32 ttfb_ms = 2;

Return values
int

getWer()

intelligibility: intended text vs a transcription of the output audio

public getWer() : float

Generated from protobuf field optional float wer = 3;

Return values
float

hasEnunciationScore()

public hasEnunciationScore() : mixed

hasNaturalnessScore()

public hasNaturalnessScore() : mixed

hasSpeechRateWpm()

public hasSpeechRateWpm() : mixed

hasTtfaMs()

public hasTtfaMs() : mixed

hasTtfbMs()

public hasTtfbMs() : mixed

hasWer()

public hasWer() : mixed

setEnunciationScore()

judged 0-1: key entities audibly intact

public setEnunciationScore(float $var) : $this

Generated from protobuf field optional float enunciation_score = 6;

Parameters
$var : float
Return values
$this

setNaturalnessScore()

judged 0-1: prosody / expressiveness

public setNaturalnessScore(float $var) : $this

Generated from protobuf field optional float naturalness_score = 5;

Parameters
$var : float
Return values
$this

setSpeechRateWpm()

speaking rate; conversational English ~110-150

public setSpeechRateWpm(float $var) : $this

Generated from protobuf field optional float speech_rate_wpm = 4;

Parameters
$var : float
Return values
$this

setTtfaMs()

Time to first AUDIO — the user-relevant signal. Some providers stream bytes before actual speech, so byte-level TTFB alone over-promises; it is kept below for disambiguation.

public setTtfaMs(int $var) : $this

Generated from protobuf field optional uint32 ttfa_ms = 1;

Parameters
$var : int
Return values
$this

setTtfbMs()

provider byte-level TTFB, agent-reported

public setTtfbMs(int $var) : $this

Generated from protobuf field optional uint32 ttfb_ms = 2;

Parameters
$var : int
Return values
$this

setWer()

intelligibility: intended text vs a transcription of the output audio

public setWer(float $var) : $this

Generated from protobuf field optional float wer = 3;

Parameters
$var : float
Return values
$this
On this page

Search results