SSML Text-to-Speech API: Premium Voices to MP3

An SSML text-to-speech API lets you mark pauses, emphasis, and say-as rules in the string, then still get an MP3 URL back. On Woord that is the same POST https://www.getwoord.com/api/convert as plain TTS: Bearer auth and form fields (text, gender_voice, language) — not a JSON body. Premium SSML voices prefix the locale with ssml_premium_ (example ssml_premium_en_US). Premium text voices use text_premium_en_US instead.

Starter is $9.99/mo (10k characters per file). Trial is 8 days / 20k characters. The Free site path does not call this API. There is no MCP. Docs: /pages/api_docs.

Use cases

• IVR prompts. Wrap the line in <speak> and set language=ssml_premium_en_US so breaks and spell-outs survive synthesis.
• E-learning. Same convert path; keep text under the 10k-character file cap.
• Locale swap. Documented locales include en_US, en_GB, es_ES, and others — prefix ssml_premium_ when the payload is SSML.

Request: premium SSML (en_US)

curl 'https://www.getwoord.com/api/convert' -X POST \
-H 'Accept: application/json' \
-H 'Authorization: Bearer YOUR_KEY' \
--data-urlencode 'text=<speak>Hello World</speak>' \
--data 'gender_voice=female' \
--data 'language=ssml_premium_en_US' \
--data 'speakingRate=1.00'

gender_voice is male, female, or neutral. Optional speakingRate is 0.25–4.0. The product wraps content in <speak> when you omit the tags.

Response: official convert sample

{
"message": "Your audio has been created!",
"audio_src": "https://getwoord.s3.amazonaws.com/4273352455515882618255eaaf3c1cbdbe0.55443890.mp3",
"error": false
}

Play or download audio_src. The S3 filename is a sample, not a stable CDN path to hotlink.

Go live

Synthesize SSML with a Woord key →