Skip to main content

TTS

Generate and manage API-created Text-to-Speech (TTS) audios, exposed through client.tts.

List TTS audios​

Returns a paginated list of API-created TTS audios belonging to your team. Audios created by a different API key for the same team are included; audios created without an API key are not.

audios = client.tts.list(
    search=None,       # search by title
    per_page=10,
    page=1,
    sort=None,           # created_at, -created_at, updated_at, -updated_at, uuid, -uuid (default -created_at)
    voice=None,           # filter by voice slug
    date_from=None,     # only audios created on/after this date
    date_to=None,        # only audios created on/before this date
    extra_params=None,   # dict of additional query parameters
    timeout=None,          # override the client's default timeout, in seconds
)

Generate a TTS audio​

Converts text into speech, charges your team for the usage, and associates the generated audio with the current API key.

audio = client.tts.generate(
    text="ދިވެހިޖީޕީޓީއަށް މަރުޙަބާ",  # up to 10,000 characters
    voice="hajja",                     # a slug returned by client.voices.list()
    extra_params=None,   # dict of additional body parameters
    timeout=None,          # override the request timeout, in seconds (defaults to 120)
)

Generation defaults to a 120 second request timeout (rather than the client's default of 30 seconds) since audio generation can take a while. Pass timeout= to override this per call.

{
    "data": {
        "uuid": "b7d7a270-7dc3-46c9-be27-d68f0ce55f36",
        "title": "ދިވެހިޖީޕީޓީއަށް މަރުޙަބާ",
        "text": "ދިވެހިޖީޕީޓީއަށް މަރުޙަބާ",
        "normalized_text": "ދިވެހިޖީޕީޓީއަށް މަރުހަބާ",
        "duration": 1860,
        "audio_url": "https://...",
        "voice": "hajja",
        "usage": {"charged_credits": 0.25, "usage_units": 25, ...},
        "created_at": "...",
        "updated_at": "...",
    }
}

Raises dhivehigpt_sdk.PermissionDeniedError when your team has no credits available, and dhivehigpt_sdk.ServerError if the speech provider fails to generate the audio.

Get a TTS audio​

audio = client.tts.get("b7d7a270-7dc3-46c9-be27-d68f0ce55f36")

Update a TTS audio​

Updates the display title of an API-created TTS audio (up to 255 characters).

audio = client.tts.update("b7d7a270-7dc3-46c9-be27-d68f0ce55f36", title="Welcome message")

update and delete also accept extra_params (body parameters, update only) and timeout keyword arguments, same as above.

Delete a TTS audio​

client.tts.delete("b7d7a270-7dc3-46c9-be27-d68f0ce55f36")

Returns None on success (HTTP 204).

All of the above raise dhivehigpt_sdk.NotFoundError when the TTS audio was not created through the API, does not exist, or does not belong to your team.