GET

/v1/schemas/elevenlabs/audio/v1/text-to-speech/{voice_id}/stream

Accept:

kind inputactivity audiohash bdd88ccd…extracted Nov 11, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `4fa3fc2e746d36f6c01d5db3e204140071b3f1b20a159ea56c21f7959c5a610f` { "fetchedAt": 1794374428, "extractorVersion": "2", "verifiedAt": null }

Request body

14 properties · 1 required · 3 $defs

property	type	description	constraints
text*	string	The text that will get converted into speech.
model_id	string	Identifier of the model that will be used, you can query them using GET /v1/models. The model needs to have support for text to speech, you can check this using the can_do_text_to_speech property.	default: "eleven_multilingual_v2"
language_code	string	Language code (ISO 639-1) used to enforce a language for the model and text normalization. If the model does not support the provided language code, it will be ignored. This parameter is not supported for multilingual_v2 models.	nullable
voice_settings	VoiceSettingsResponseModel	Voice settings overriding stored settings for the given voice. They are applied only on the given request.	nullable
pronunciation_dictionary_locators	PronunciationDictionaryVersionLocatorRequestModel[]	A list of pronunciation dictionary locators (id, version_id) to be applied to the text. They will be applied in order. You may have up to 3 locators per request	nullable
seed	integer	If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result. Determinism is not guaranteed. Must be integer between 0 and 4294967295.	nullable
previous_text	string	The text that came before the text of the current request. Can be used to improve the speech's continuity when concatenating together multiple generations or to influence the speech's continuity in the current generation.	nullable
next_text	string	The text that comes after the text of the current request. Can be used to improve the speech's continuity when concatenating together multiple generations or to influence the speech's continuity in the current generation.	nullable
previous_request_ids	string[]	A list of request_id of the samples that were generated before this generation. Can be used to improve the speech's continuity when splitting up a large task into multiple requests. The results will be best when the same model is used across the generations. In case both previous_text and previous_request_ids is send, previous_text will be ignored. A maximum of 3 request_ids can be sent.	nullable
next_request_ids	string[]	A list of request_id of the samples that come after this generation. next_request_ids is especially useful for maintaining the speech's continuity when regenerating a sample that has had some audio quality issues. For example, if you have generated 3 speech clips, and you want to improve clip 2, passing the request id of clip 3 as a next_request_id (and that of clip 1 as a previous_request_id) will help maintain natural flow in the combined speech. The results will be best when the same model is used across the generations. In case both next_text and next_request_ids is send, next_text will be ignored. A maximum of 3 request_ids can be sent.	nullable
use_pvc_as_ivc	boolean	Whether to use the IVC version of a professional voice. This may improve expressiveness and reduce latency.	nullable
apply_text_normalization	enum	This parameter controls text normalization with three modes: 'auto', 'on', and 'off'. When set to 'auto', the system will automatically decide whether to apply text normalization (e.g., spelling out numbers). With 'on', text normalization will always be applie…	auto \| on \| off · default: "auto"
apply_language_text_normalization	boolean	This parameter controls language text normalization. This helps with proper pronunciation of text in some supported languages. WARNING: This parameter can heavily increase the latency of the request. Currently only supported for Japanese.	default: false · deprecated
audio_effects	EffectsSpec-Input	Audio effects applied to generated speech: a filter preset, perceived distance, a reverb environment, a background noise bed, and stereo panning. Leave every field at its default to get unprocessed audio.	nullable

Types

3 $defs

EffectsSpec-Inputobject

Audio effects applied to generated speech: a filter preset, perceived distance, a reverb environment, a background noise bed, and stereo panning. Leave every field at its default to get unprocessed audio.

filter_preset_id	string	ID of a filter preset that colors the voice, such as `phone` or `old_radio`. List valid IDs with `GET /v1/audio-effects/catalog`.	nullable
distance	number	How far the speaker sounds from the microphone, from 0 (close, unchanged) to 1 (far). Higher values roll off low and high frequencies and, when an environment is set, shift the balance from the dry voice toward the reverb.	0 ≤ n ≤ 1 · default: 0
environment_id	string	ID of a reverb environment that places the voice in a space, such as `small_room` or `hall`. List valid IDs with `GET /v1/audio-effects/catalog`. Reverb is rendered in stereo and adds a tail of up to several seconds after the speech ends.	nullable
background_noise_id	string	ID of a background noise bed looped under the voice, such as `cafe` or `keyboard`. List valid IDs with `GET /v1/audio-effects/catalog`.	nullable
send_level	number	How much of the voice feeds the reverb, from 0 (none) to 1 (full). Only applies when `environment_id` is set; 0 turns the reverb off.	0 ≤ n ≤ 1 · default: 1
pan	number	Stereo position of the voice, from -1 (hard left) through 0 (center) to 1 (hard right). Only the voice is panned; reverb and background noise stay centered.	-1 ≤ n ≤ 1 · default: 0

PronunciationDictionaryVersionLocatorRequestModelobject

pronunciation_dictionary_id*	string	The ID of the pronunciation dictionary.
version_id	string	The ID of the version of the pronunciation dictionary. If not provided, the latest version will be used.	nullable

VoiceSettingsResponseModelobject

stability	number	Determines how stable the voice is and the randomness between each generation. Lower values introduce broader emotional range for the voice. Higher values can result in a monotonous voice with limited emotion.	0 ≤ n ≤ 1 · nullable
use_speaker_boost	boolean	This setting boosts the similarity to the original speaker. Using this setting requires a slightly higher computational load, which in turn increases latency.	nullable
similarity_boost	number	Determines how closely the AI should adhere to the original voice when attempting to replicate it.	0 ≤ n ≤ 1 · nullable
style	number	Determines the style exaggeration of the voice. This setting attempts to amplify the style of the original speaker. It does consume additional computational resources and might increase latency if set to anything other than 0.	nullable
speed	number	Adjusts the speed of the voice. A value of 1.0 is the default speed, while values less than 1.0 slow down the speech, and values greater than 1.0 speed it up.	nullable
clone_strength	number	Controls how closely the output matches your professional voice clone samples (0–1). Lower values sound more versatile, while higher values preserve more of the original voice characteristics. We recommend 0 for most use cases, but if you do run into issues 0.5 could be a good fallback. Applies only to professional voice clones fine-tuned for the selected v4 model.	0 ≤ n ≤ 1 · nullable

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"elevenlabs","endpointId":"v1/text-to-speech/{voice_id}/stream","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}