Models

Protocol "provider": "sarvam"LiveKit plugin tcr.TTS(model="sarvam/…")

Models

bulbul:v3, bulbul:v4-flash

Languages

11

Speed

0.5 – 2.0

Sample rates

8, 16, 22.05, 24, 32, 44.1, 48 kHz

Formats

pcm, wav, mp3, aac, mulaw, alaw

Models and voices#

Languages#

• Banglabn-IN
• Englishen-IN
• Gujaratigu-IN
• Hindihi-IN
• Kannadakn-IN
• Malayalamml-IN
• Marathimr-IN
• Odiaod-IN
• Punjabipa-IN
• Tamilta-IN
• Telugute-IN

Settings#

Fields of the start message (top-level fields on /v1/audio/speech). The LiveKit plugin’s TTS and the Pipecat service’s Settings take them in the same names, with model as "sarvam/<model>" and the options in options.

modelstringRequired

One of bulbul:v3, bulbul:v4-flash

Which model synthesizes the segment.

voicestringRequired

A voice of the chosen model, passed to Sarvam as given. See models and voices.

languagestringRequired

A Sarvam language code from languages, passed through as given.

speednumberDefault 1.0

Range 0.5 – 2.0

Speech speed. 1.0 is normal speed.

sample_rateintegerRequired

One of 8000, 16000, 22050, 24000

Of the audio you receive, in Hz. /v1/audio/speech also takes 32000, 44100, 48000.

formatstringDefault "pcm"

One of pcm, wav, mp3, aac, mulaw, alaw

The audio format, produced by Sarvam itself. wav is response_format on /v1/audio/speech only.

options.temperaturenumberDefault 0.6

Range 0.01 – 1.0

Sampling temperature: lower is steadier, higher is more expressive.

options.dict_idstring | nullDefault null

A Sarvam pronunciation dictionary to apply.

options.bitratestringDefault "128k"

One of 32k, 64k, 96k, 128k, 192k

Bitrate of mp3 and aac audio.

options.pitchnumberbulbul:v4-flash only. Default 0.0

Range -0.5 – 0.5

Pitch: lower is deeper, higher is sharper.

options.loudnessnumberbulbul:v4-flash only. Default 1.0

Range 0.1 – 2.5

Loudness: lower is quieter, higher is louder.

options.enable_preprocessingbooleanbulbul:v4-flash only. Default false

Normalize numbers and English words in the text before synthesis.

options.min_buffer_sizeinteger | nullDefault null

Range 30 – 200

Characters Sarvam buffers before it starts synthesizing. Null: the router's choice, the least Sarvam allows.

options.max_chunk_lengthinteger | nullDefault null

Range 50 – 500

The longest text Sarvam synthesizes in one piece; it splits longer sentences. Null: the router's choice, the most Sarvam allows, so each sentence is synthesized whole.

from livekit.plugins import tts_cache_router as tcrtts = tcr.TTS(    model="sarvam/bulbul:v3",    voice="shubh",    language="bn-IN",    speed=1.0,    options={"temperature": 0.6, "dict_id": None, "min_buffer_size": None, "max_chunk_length": None},    sample_rate=24000,)

{  "type": "start",  "segment_id": "c0ffee",  "provider": "sarvam",  "model": "bulbul:v3",  "voice": "shubh",  "language": "bn-IN",  "speed": 1.0,  "sample_rate": 24000,  "format": "pcm",  "options": {"temperature": 0.6, "dict_id": null, "min_buffer_size": null, "max_chunk_length": null}}

Protocol "provider": "raya"LiveKit plugin tcr.TTS(model="raya/…")

Models

m1, standard

Languages

14

Speed

0.5 – 1.5

Sample rates

8, 16, 22.05, 24 kHz

Formats

pcm, wav, mulaw

Models and voices#

Languages#

• Hindihi
• Marathimr
• Telugute
• Kannadakn
• Banglabn
• Assameseas
• Gujaratigu
• Nepaline
• Malayalamml
• Tamilta
• English (India)en-in
• English (United States)en-us
• Englishm1 onlyen
• Punjabim1 onlypa

Settings#

Fields of the start message (top-level fields on /v1/audio/speech). The LiveKit plugin’s TTS and the Pipecat service’s Settings take them in the same names, with model as "raya/<model>" and the options in options. Raya takes no options: send {} or leave it out.

modelstringRequired

One of m1, standard

Which model synthesizes the segment.

voicestringRequired

A voice of the chosen model, passed to Raya as given. See models and voices.

languagestringRequired

A Raya language code from languages, passed through as given.

speednumberDefault 1.0

Range 0.5 – 1.5

Speech speed. 1.0 is normal speed.

sample_rateintegerRequired

One of 8000, 16000, 22050, 24000

Of the audio you receive, in Hz.

formatstringDefault "pcm"

One of pcm, wav, mulaw

The audio format, produced by Raya itself. wav is response_format on /v1/audio/speech only.

from livekit.plugins import tts_cache_router as tcrtts = tcr.TTS(    model="raya/m1",    voice="<voice id>",    language="hi",    speed=1.0,    sample_rate=24000,)

{  "type": "start",  "segment_id": "c0ffee",  "provider": "raya",  "model": "m1",  "voice": "<voice id>",  "language": "hi",  "speed": 1.0,  "sample_rate": 24000,  "format": "pcm",  "options": {}}

Protocol "provider": "elevenlabs"LiveKit plugin tcr.TTS(model="elevenlabs/…")

Models

eleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_flash_v2, eleven_turbo_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo

Languages

94

Speed

set in voice_settings

Sample rates

8, 16, 22.05, 24, 32, 44.1, 48 kHz

Formats

pcm, wav, mp3, mulaw, alaw

Models and voices#

Languages#

• Englishen
• Japaneseeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyja
• Chineseeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyzh
• Germaneleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyde
• Hindieleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyhi
• Frencheleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyfr
• Koreaneleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyko
• Portugueseeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlypt
• Italianeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyit
• Spanisheleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyes
• Russianeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyru
• Indonesianeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyid
• Dutcheleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlynl
• Turkisheleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlytr
• Filipinoeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyfil
• Polisheleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlypl
• Swedisheleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlysv
• Bulgarianeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlybg
• Romanianeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyro
• Arabiceleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyar
• Czecheleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlycs
• Greekeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyel
• Finnisheleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyfi
• Croatianeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyhr
• Malayeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyms
• Slovakeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlysk
• Danisheleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyda
• Tamileleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyta
• Ukrainianeleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyuk
• Hungarianeleven_flash_v2_5, eleven_turbo_v2_5, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyhu
• Norwegianeleven_flash_v2_5, eleven_turbo_v2_5, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyno
• Vietnameseeleven_flash_v2_5, eleven_turbo_v2_5, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyvi
• Afrikaanseleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyaf
• Armenianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyhy
• Assameseeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyas
• Azerbaijanieleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyaz
• Belarusianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlybe
• Banglaeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlybn
• Bosnianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlybs
• Catalaneleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyca
• Cebuanoeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyceb
• Nyanjaeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyny
• Estonianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyet
• Galicianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlygl
• Georgianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyka
• Gujaratieleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlygu
• Hausaeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyha
• Hebreweleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyhe
• Icelandiceleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyis
• Irisheleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyga
• Javaneseeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyjv
• Kannadaeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlykn
• Kazakheleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlykk
• Kyrgyzeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyky
• Latvianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlylv
• Lingalaeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyln
• Lithuanianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlylt
• Luxembourgisheleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlylb
• Macedonianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlymk
• Malayalameleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyml
• Marathieleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlymr
• Nepalieleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyne
• Pashtoeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyps
• Persianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyfa
• Punjabieleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlypa
• Serbianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlysr
• Sindhieleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlysd
• Slovenianeleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlysl
• Somalieleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyso
• Swahilieleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlysw
• Telugueleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyte
• Thaieleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyth
• Urdueleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlyur
• Welsheleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo onlycy
• Amhariceleven_v4, eleven_v4_turbo onlyam
• Asturianeleven_v4, eleven_v4_turbo onlyast
• Burmeseeleven_v4, eleven_v4_turbo onlymy
• Cantoneseeleven_v4, eleven_v4_turbo onlyyue
• Fulaeleven_v4, eleven_v4_turbo onlyff
• Kambaeleven_v4, eleven_v4_turbo onlykam
• Laoeleven_v4, eleven_v4_turbo onlylo
• Gandaeleven_v4, eleven_v4_turbo onlylg
• Malteseeleven_v4, eleven_v4_turbo onlymt
• Māorieleven_v4, eleven_v4_turbo onlymi
• Mongolianeleven_v4, eleven_v4_turbo onlymn
• Occitaneleven_v4, eleven_v4_turbo onlyoc
• Odiaeleven_v4, eleven_v4_turbo onlyor
• Shonaeleven_v4, eleven_v4_turbo onlysn
• Central Kurdisheleven_v4, eleven_v4_turbo onlyckb
• Tajikeleven_v4, eleven_v4_turbo onlytg
• Uzbekeleven_v4, eleven_v4_turbo onlyuz
• Wolofeleven_v4, eleven_v4_turbo onlywo
• Yorubaeleven_v4, eleven_v4_turbo onlyyo
• Zulueleven_v4, eleven_v4_turbo onlyzu

Settings#

Fields of the start message (top-level fields on /v1/audio/speech). The LiveKit plugin’s TTS and the Pipecat service’s Settings take them in the same names, with model as "elevenlabs/<model>" and the options in options.

modelstringRequired

One of eleven_flash_v2_5, eleven_turbo_v2_5, eleven_multilingual_v2, eleven_flash_v2, eleven_turbo_v2, eleven_v3, eleven_v3_conversational, eleven_v4, eleven_v4_turbo

Which model synthesizes the segment.

voicestringRequired

A voice of the chosen model, passed to ElevenLabs as given. See models and voices.

languagestring | nullOptional

A ElevenLabs language code from languages, passed through as given. Leave it out and ElevenLabs detects the language.

sample_rateintegerRequired

One of 8000, 16000, 22050, 24000, 32000, 44100, 48000

Of the audio you receive, in Hz.

formatstringDefault "pcm"

One of pcm, wav, mp3, mulaw, alaw

The audio format, produced by ElevenLabs itself. wav is response_format on /v1/audio/speech only.

options.voice_settingsobject | nullDefault null

Any of stability, similarity_boost and style (0–1), speed (0.7–1.2) and use_speaker_boost; one left out keeps the voice's stored setting. Null: the voice's stored settings. Dialogue models (eleven_v3, eleven_v4) use stability only.

options.apply_text_normalizationstringDefault "auto"

One of auto, on, off

Spell out numbers, dates and the like before speaking.

options.apply_language_text_normalizationboolean | nullDefault null

Language-specific normalization (Japanese); adds latency.

options.enable_ssml_parsingbooleanDefault false

Parse SSML tags in the text.

options.enable_loggingbooleanDefault true

false: ElevenLabs' zero-retention mode (enterprise accounts).

options.pronunciation_dictionary_locatorsarray | nullDefault null

Up to 3 {pronunciation_dictionary_id, version_id}.

options.bitratestring | nullDefault null

One of 32k, 48k, 64k, 96k, 128k, 192k

Bitrate of mp3: 32k at 22050 Hz, 48k at 24000 Hz, 32k–192k at 44100 Hz (default 128k).

options.sync_alignmentbooleanDefault true

/v1/stream: send each sentence's word timings (words messages).

options.preferred_alignmentstring | nullDefault null

One of normalized, original

/v1/stream: words of the text as sent (original) or as ElevenLabs normalized it. Null: original for Japanese, Korean and Chinese, else normalized.

from livekit.plugins import tts_cache_router as tcrtts = tcr.TTS(    model="elevenlabs/eleven_flash_v2_5",    voice="<voice id>",    options={"apply_text_normalization": "auto", "apply_language_text_normalization": None, "enable_ssml_parsing": False, "enable_logging": True, "sync_alignment": True, "preferred_alignment": None},    sample_rate=24000,)

{  "type": "start",  "segment_id": "c0ffee",  "provider": "elevenlabs",  "model": "eleven_flash_v2_5",  "voice": "<voice id>",  "sample_rate": 24000,  "format": "pcm",  "options": {"voice_settings": null, "apply_text_normalization": "auto", "apply_language_text_normalization": null, "enable_ssml_parsing": false, "enable_logging": true, "pronunciation_dictionary_locators": null, "sync_alignment": true, "preferred_alignment": null}}

Protocol "provider": "openai"LiveKit plugin tcr.TTS(model="openai/…")

Models

gpt-4o-mini-tts, gpt-4o-mini-tts-2025-12-15, tts-1, tts-1-hd

Languages

57

Speed

0.25 – 4.0

Sample rates

24 kHz

Formats

pcm, wav, mp3, aac

Models and voices#

Languages#

OpenAI takes no language setting: it speaks the language of the text.

• Afrikaansaf
• Arabicar
• Armenianhy
• Azerbaijaniaz
• Belarusianbe
• Bosnianbs
• Bulgarianbg
• Catalanca
• Chinesezh
• Croatianhr
• Czechcs
• Danishda
• Dutchnl
• Englishen
• Estonianet
• Finnishfi
• Frenchfr
• Galiciangl
• Germande
• Greekel
• Hebrewhe
• Hindihi
• Hungarianhu
• Icelandicis
• Indonesianid
• Italianit
• Japaneseja
• Kannadakn
• Kazakhkk
• Koreanko
• Latvianlv
• Lithuanianlt
• Macedonianmk
• Malayms
• Marathimr
• Māorimi
• Nepaline
• Norwegianno
• Persianfa
• Polishpl
• Portuguesept
• Romanianro
• Russianru
• Serbiansr
• Slovaksk
• Sloveniansl
• Spanishes
• Swahilisw
• Swedishsv
• Filipinotl
• Tamilta
• Thaith
• Turkishtr
• Ukrainianuk
• Urduur
• Vietnamesevi
• Welshcy

Settings#

Fields of the start message (top-level fields on /v1/audio/speech). The LiveKit plugin’s TTS and the Pipecat service’s Settings take them in the same names, with model as "openai/<model>" and the options in options.

modelstringRequired

One of gpt-4o-mini-tts, gpt-4o-mini-tts-2025-12-15, tts-1, tts-1-hd

Which model synthesizes the segment.

voicestringRequired

A voice of the chosen model, passed to OpenAI as given. See models and voices.

speednumberDefault 1.0

Range 0.25 – 4.0

Speech speed. 1.0 is normal speed.

sample_rateintegerRequired

One of 24000

Of the audio you receive, in Hz.

formatstringDefault "pcm"

One of pcm, wav, mp3, aac

The audio format, produced by OpenAI itself. wav is response_format on /v1/audio/speech only.

options.instructionsstring | nullDefault null

How to speak: tone, accent, emotion, pace. gpt-4o-mini-tts follows them; tts-1 and tts-1-hd ignore them.

from livekit.plugins import tts_cache_router as tcrtts = tcr.TTS(    model="openai/gpt-4o-mini-tts",    voice="alloy",    speed=1.0,    options={"instructions": None},    sample_rate=24000,)

{  "type": "start",  "segment_id": "c0ffee",  "provider": "openai",  "model": "gpt-4o-mini-tts",  "voice": "alloy",  "speed": 1.0,  "sample_rate": 24000,  "format": "pcm",  "options": {"instructions": null}}

Protocol "provider": "deepgram"LiveKit plugin tcr.TTS(model="deepgram/…")

Models

aura-2, aura-1, flux

Languages

7

Speed

0.7 – 1.5

Sample rates

8, 16, 22.05, 24, 32, 44.1, 48 kHz

Formats

pcm, wav, mulaw, alaw, mp3

Models and voices#

Languages#

Deepgram takes no language setting: it speaks the language of the text.

• Englishen
• Spanishaura-2 onlyes
• Germanaura-2 onlyde
• Frenchaura-2 onlyfr
• Italianaura-2 onlyit
• Japaneseaura-2 onlyja
• Dutchaura-2 onlynl

Settings#

Fields of the start message (top-level fields on /v1/audio/speech). The LiveKit plugin’s TTS and the Pipecat service’s Settings take them in the same names, with model as "deepgram/<model>" and the options in options.

modelstringRequired

One of aura-2, aura-1, flux

Which model synthesizes the segment.

voicestringRequired

A voice of the chosen model, passed to Deepgram as given. See models and voices.

speednumberOptional

Range 0.7 – 1.5

Speech speed, aura-2 only. 1.0 is normal speed; left out, the provider's default.

sample_rateintegerRequired

One of 8000, 16000, 22050, 24000, 32000, 44100, 48000

Of the audio you receive, in Hz.

formatstringDefault "pcm"

One of pcm, wav, mulaw, alaw, mp3

The audio format, produced by Deepgram itself. wav is response_format on /v1/audio/speech only.

options.mip_opt_outbooleanDefault false

Opt out of Deepgram's Model Improvement Program.

options.bit_rateinteger | nullaura-2, aura-1 only. Default null

Range 32000 – 48000

MP3's bit rate in bits per second: 32000 or 48000. Null: Deepgram's default (48000).

from livekit.plugins import tts_cache_router as tcrtts = tcr.TTS(    model="deepgram/aura-2",    voice="agathe-fr",    options={"mip_opt_out": False},    sample_rate=24000,)

{  "type": "start",  "segment_id": "c0ffee",  "provider": "deepgram",  "model": "aura-2",  "voice": "agathe-fr",  "sample_rate": 24000,  "format": "pcm",  "options": {"mip_opt_out": false}}

Protocol "provider": "xai"LiveKit plugin tcr.TTS(model="xai/…")

Models

xai-tts

Languages

21

Speed

0.7 – 1.5

Sample rates

8, 16, 22.05, 24, 44.1, 48 kHz

Formats

pcm, wav, mp3, mulaw, alaw

Models and voices#

Languages#

• Detectedauto
• Englishen
• Arabic (Egypt)ar-EG
• Arabic (Saudi Arabia)ar-SA
• Arabic (United Arab Emirates)ar-AE
• Banglabn
• Chinesezh
• Frenchfr
• Germande
• Hindihi
• Indonesianid
• Italianit
• Japaneseja
• Koreanko
• Portuguese (Brazil)pt-BR
• Portuguese (Portugal)pt-PT
• Russianru
• Spanish (Mexico)es-MX
• Spanish (Spain)es-ES
• Turkishtr
• Vietnamesevi

Settings#

Fields of the start message (top-level fields on /v1/audio/speech). The LiveKit plugin’s TTS and the Pipecat service’s Settings take them in the same names, with model as "xai/<model>" and the options in options.

modelstringRequired

One of xai-tts

Which model synthesizes the segment.

voicestringRequired

A voice of the chosen model, passed to xAI as given. See models and voices.

languagestring | nullOptional

A xAI language code from languages, passed through as given. Leave it out and xAI detects the language.

speednumberOptional

Range 0.7 – 1.5

Speech speed. 1.0 is normal speed; left out, the provider's default.

sample_rateintegerRequired

One of 8000, 16000, 22050, 24000, 44100, 48000

Of the audio you receive, in Hz.

formatstringDefault "pcm"

One of pcm, wav, mp3, mulaw, alaw

The audio format, produced by xAI itself. wav is response_format on /v1/audio/speech only.

options.optimize_streaming_latencyinteger | nullDefault null

Range 0 – 2

0: best quality (xAI's default); 1 and 2: a smaller first chunk for lower time to first audio, with a quality tradeoff at chunk boundaries. Null: xAI's default.

options.text_normalizationboolean | nullDefault null

Normalize written forms (numbers, abbreviations, symbols) into spoken forms before synthesis. Null: xAI's default (false).

options.bit_rateinteger | nullDefault null

Range 32000 – 192000

MP3's bit rate in bits per second: 32000, 64000, 96000, 128000 or 192000. Null: xAI's default (128000).

from livekit.plugins import tts_cache_router as tcrtts = tcr.TTS(    model="xai/xai-tts",    voice="altair",    options={"optimize_streaming_latency": None, "text_normalization": None},    sample_rate=24000,)

{  "type": "start",  "segment_id": "c0ffee",  "provider": "xai",  "model": "xai-tts",  "voice": "altair",  "sample_rate": 24000,  "format": "pcm",  "options": {"optimize_streaming_latency": null, "text_normalization": null}}

Protocol "provider": "soniox"LiveKit plugin tcr.TTS(model="soniox/…")

Models

tts-rt-v2

Languages

63

Speed

0.7 – 1.3

Sample rates

8, 16, 24, 32, 44.1, 48 kHz

Formats

pcm, wav, mp3, mulaw, alaw

Models and voices#

Languages#

• Afrikaansaf
• Arabicar
• Azerbaijaniaz
• Belarusianbe
• Bulgarianbg
• Banglabn
• Bosnianbs
• Catalanca
• Czechcs
• Welshcy
• Danishda
• Germande
• Greekel
• Englishen
• Spanishes
• Estonianet
• Basqueeu
• Persianfa
• Finnishfi
• Frenchfr
• Galiciangl
• Gujaratigu
• Hebrewhe
• Hindihi
• Croatianhr
• Hungarianhu
• Indonesianid
• Icelandicis
• Italianit
• Japaneseja
• Kazakhkk
• Kannadakn
• Koreanko
• Lithuanianlt
• Latvianlv
• Macedonianmk
• Malayalamml
• Marathimr
• Malayms
• Dutchnl
• Norwegianno
• Punjabipa
• Polishpl
• Portuguesept
• Romanianro
• Russianru
• Slovaksk
• Sloveniansl
• Albaniansq
• Serbiansr
• Sundanesesu
• Swedishsv
• Swahilisw
• Tamilta
• Telugute
• Thaith
• Filipinotl
• Turkishtr
• Ukrainianuk
• Urduur
• Uzbekuz
• Vietnamesevi
• Chinesezh

Settings#

Fields of the start message (top-level fields on /v1/audio/speech). The LiveKit plugin’s TTS and the Pipecat service’s Settings take them in the same names, with model as "soniox/<model>" and the options in options.

modelstringRequired

One of tts-rt-v2

Which model synthesizes the segment.

voicestringRequired

A voice of the chosen model, passed to Soniox as given. See models and voices.

languagestringRequired

A Soniox language code from languages, passed through as given.

speednumberDefault 1.0

Range 0.7 – 1.3

Speech speed. 1.0 is normal speed.

sample_rateintegerRequired

One of 8000, 16000, 24000, 32000, 44100, 48000

Of the audio you receive, in Hz.

formatstringDefault "pcm"

One of pcm, wav, mp3, mulaw, alaw

The audio format, produced by Soniox itself. wav is response_format on /v1/audio/speech only.

options.bitrateinteger | nullDefault null

Range 32000 – 320000

MP3's bit rate in bits per second: 32000, 64000, 96000, 128000, 192000, 256000 or 320000. Null: Soniox's default (128000).

options.reduce_silenceboolean | nullDefault null

Shorten the pauses between words. Null: Soniox's default (false).

from livekit.plugins import tts_cache_router as tcrtts = tcr.TTS(    model="soniox/tts-rt-v2",    voice="Daniel",    language="af",    speed=1.0,    options={"reduce_silence": None},    sample_rate=24000,)

{  "type": "start",  "segment_id": "c0ffee",  "provider": "soniox",  "model": "tts-rt-v2",  "voice": "Daniel",  "language": "af",  "speed": 1.0,  "sample_rate": 24000,  "format": "pcm",  "options": {"reduce_silence": null}}

Protocol "provider": "smallestai"LiveKit plugin tcr.TTS(model="smallestai/…")

Models

lightning_v3.1, lightning_v3.1_pro

Languages

33

Speed

0.5 – 2.0

Sample rates

8, 16, 24, 44.1 kHz

Formats

pcm, wav, mp3, mulaw, alaw

Models and voices#

Languages#

• Detectedauto
• Englishen
• Hindihi
• Marathimr
• Kannadakn
• Tamilta
• Banglabn
• Gujaratigu
• Germande
• Frenchfr
• Spanishes
• Italianit
• Polishpl
• Dutchnl
• Russianru
• Arabicar
• Hebrewlightning_v3.1 onlyhe
• Swedishsv
• Malayalamml
• Telugute
• Portuguesept
• Punjabipa
• Odiaor
• Greeklightning_v3.1_pro onlyel
• Finnishlightning_v3.1_pro onlyfi
• Norwegianlightning_v3.1_pro onlyno
• Chineselightning_v3.1_pro onlyzh
• Indonesianlightning_v3.1_pro onlyid
• Japaneselightning_v3.1_pro onlyja
• Koreanlightning_v3.1_pro onlyko
• Malaylightning_v3.1_pro onlyms
• Turkishlightning_v3.1_pro onlytr
• Vietnameselightning_v3.1_pro onlyvi

Settings#

Fields of the start message (top-level fields on /v1/audio/speech). The LiveKit plugin’s TTS and the Pipecat service’s Settings take them in the same names, with model as "smallestai/<model>" and the options in options. Smallest AI takes no options: send {} or leave it out.

modelstringRequired

One of lightning_v3.1, lightning_v3.1_pro

Which model synthesizes the segment.

voicestringRequired

A voice of the chosen model, passed to Smallest AI as given. See models and voices.

languagestringDefault "en"

A Smallest AI language code from languages, passed through as given.

speednumberDefault 1.0

Range 0.5 – 2.0

Speech speed. 1.0 is normal speed.

sample_rateintegerRequired

One of 8000, 16000, 24000, 44100

Of the audio you receive, in Hz.

formatstringDefault "pcm"

One of pcm, wav, mp3, mulaw, alaw

The audio format, produced by Smallest AI itself. wav is response_format on /v1/audio/speech only.

from livekit.plugins import tts_cache_router as tcrtts = tcr.TTS(    model="smallestai/lightning_v3.1",    voice="lakshya",    language="en",    speed=1.0,    sample_rate=24000,)

{  "type": "start",  "segment_id": "c0ffee",  "provider": "smallestai",  "model": "lightning_v3.1",  "voice": "lakshya",  "language": "en",  "speed": 1.0,  "sample_rate": 24000,  "format": "pcm",  "options": {}}

Protocol "provider": "cartesia"LiveKit plugin tcr.TTS(model="cartesia/…")

Models

sonic-3.6, sonic-3.5, sonic-3

Languages

44

Speed

0.6 – 1.5

Sample rates

8, 16, 22.05, 24, 44.1, 48 kHz

Formats

pcm, wav, mp3, mulaw, alaw

Models and voices#

Languages#

• Englishen
• Frenchfr
• Germande
• Spanishes
• Portuguesept
• Chinesezh
• Japaneseja
• Hindihi
• Italianit
• Koreanko
• Dutchnl
• Polishpl
• Russianru
• Swedishsv
• Turkishtr
• Filipinotl
• Bulgarianbg
• Romanianro
• Arabicar
• Czechcs
• Greekel
• Finnishfi
• Croatianhr
• Malayms
• Slovaksk
• Danishda
• Tamilta
• Ukrainianuk
• Hungarianhu
• Norwegianno
• Vietnamesevi
• Banglabn
• Thaith
• Hebrewhe
• Georgianka
• Indonesianid
• Telugute
• Gujaratigu
• Kannadakn
• Malayalamml
• Marathimr
• Punjabipa
• Odiasonic-3.6 onlyor
• Urdusonic-3.6 onlyur

Settings#

Fields of the start message (top-level fields on /v1/audio/speech). The LiveKit plugin’s TTS and the Pipecat service’s Settings take them in the same names, with model as "cartesia/<model>" and the options in options.

modelstringRequired

One of sonic-3.6, sonic-3.5, sonic-3

Which model synthesizes the segment.

voicestringRequired

A voice of the chosen model, passed to Cartesia as given. See models and voices.

languagestringDefault "en"

A Cartesia language code from languages, passed through as given.

speednumberOptional

Range 0.6 – 1.5

Speech speed. 1.0 is normal speed; left out, the provider's default.

sample_rateintegerRequired

One of 8000, 16000, 22050, 24000, 44100, 48000

Of the audio you receive, in Hz.

formatstringDefault "pcm"

One of pcm, wav, mp3, mulaw, alaw

The audio format, produced by Cartesia itself. wav is response_format on /v1/audio/speech only.

options.volumenumber | nullDefault null

Range 0.5 – 2.0

Loudness relative to the voice's own. Null: Cartesia's default (1.0).

options.emotionstring | nullDefault null

Guide the delivery (English): neutral, calm, angry, content, sad, scared, and the others Cartesia lists. Null: Cartesia reads the emotion from the text.

options.pronunciation_dict_idstring | nullDefault null

A Cartesia pronunciation dictionary's id. Cached takes keep the dictionary as it was when they were synthesized, until they expire.

options.bit_rateinteger | nullDefault 128000

Range 32000 – 192000

MP3's bit rate in bits per second: 32000, 64000, 96000, 128000 or 192000.

options.word_timestampsbooleanDefault true

/v1/stream: send each sentence's word timings (words messages). Not with mp3.

from livekit.plugins import tts_cache_router as tcrtts = tcr.TTS(    model="cartesia/sonic-3.6",    voice="<voice id>",    language="en",    options={"volume": None, "emotion": None, "pronunciation_dict_id": None, "word_timestamps": True},    sample_rate=24000,)

{  "type": "start",  "segment_id": "c0ffee",  "provider": "cartesia",  "model": "sonic-3.6",  "voice": "<voice id>",  "language": "en",  "sample_rate": 24000,  "format": "pcm",  "options": {"volume": null, "emotion": null, "pronunciation_dict_id": null, "word_timestamps": true}}