Skip to main content
GET
cURL

Continue building

Get a voice

Inspect a selected voice before synthesis.

List Voice Metadata options

Discover valid filter and metadata values.

Convert text to speech

Generate speech with the selected voice.

Authorizations

xi-api-key
string
header
required

Breeze Developer API key.

Query Parameters

search
string | null

Search public catalog voices by semantic and text relevance, and personal voices by name, description, or voice identifier.

language_code
enum<string>

Only return voices whose saved audio language matches this supported ISO 639-1 two-letter language code. See supported language codes.

Available options:
af,
ar,
bg,
bn,
bs,
ca,
cs,
da,
de,
el,
en,
es,
et,
eu,
fa,
fi,
fr,
gl,
he,
hi,
hu,
id,
is,
it,
ja,
kn,
ko,
lt,
mr,
ms,
ne,
nl,
no,
pl,
pt,
ro,
ru,
sk,
sl,
sq,
sv,
sw,
ta,
te,
th,
tl,
tr,
uk,
ur,
vi,
zh
primary_category_code
enum<string> | null

Only return voices in this exact primary category. See category codes.

Available options:
narration,
social-media,
ads-brand,
podcast,
learning,
support-agents,
roleplay,
gaming,
animation,
cinematic,
wellness,
stylized
gender
enum<string>[] | null

Repeat to match any supplied gender code. See Metadata options.

Available options:
male,
female,
neutral
age
enum<string>[] | null

Repeat to match any supplied age code. See Metadata options.

Available options:
child,
young,
middle_aged,
old
tone
enum<string>[] | null

Repeat to match voices containing any supplied tone code. See Metadata options.

Available options:
warm,
calm,
bright,
gentle,
energetic,
authoritative,
sincere,
weary,
precise,
refined,
urgent,
friendly,
articulate,
steady,
playful,
compassionate,
reflective,
measured,
rhythmic,
passionate
accent
enum<string> | null

Filter by accent code. american also matches English voices whose accent is null. Cannot be combined with accent_mode=unmarked. See Metadata options, accent codes.

Available options:
american,
british,
scottish,
irish,
australian,
canadian,
us_southern,
us_new_york,
indian,
south_african,
russian,
japanese,
korean,
chinese,
mandarin_northeastern,
mandarin_shaanxi,
mandarin_shanghai,
mandarin_sichuan,
mandarin_henan,
mandarin_beijing,
mandarin_taiwan,
cantonese,
mandarin_guangdong,
mandarin_yunnan
accent_mode
enum<string>
default:all

all (default) applies no null-only filter. unmarked returns voices whose accent is null; combine with language_code=zh to find Mandarin voices. Cannot be combined with accent.

Available options:
all,
unmarked
origin
string | null

Optional origin filter. Accepted values: designed, cloned.

voice_type
string
default:all

Voice source filter. Accepted values: all, default, personal.

sort
string
default:created_at_unix

Sort field for non-search listings. Accepted values: created_at_unix, name, trend. trend is available only with voice_type=default. Eligible voices absent from the daily snapshot follow ranked voices; pagination pins the snapshot and initial supplemental candidates.

sort_direction
string
default:desc

Sort direction for non-search listings. Accepted values: asc, desc. The default is desc. For trend, desc means most popular first.

next_page_token
string | null

Pagination token from the previous response. Keep the same filters.

page
integer
default:1

Page number for offset-based navigation.

Required range: x >= 1
page_size
integer
default:50

Number of records to return per page.

Required range: 1 <= x <= 100
favorites_only
boolean
default:false

Only return voices favorited by the authenticated account.

tags
string[] | null

Return voices matching all supplied tags, ignoring case. See Tags and favorites.

Response

Paginated voice list.

voices
VoiceResponse · object[]
required
has_more
boolean
required
total
integer
required
page
integer
required
page_size
integer
required
next_page_token
string | null