Changelog

simba-3.5, simba-3.2 and simba-3.0 now apply <prosody pitch> and <prosody volume>, which they used to accept without applying. They take the values Simba 1.6 takes, on POST /v1/audio/speech, POST /v1/audio/stream and POST /v1/audio/stream/with-timestamps.

• Percent volume is relative. volume="+20%" is 1.2 times the voice’s own loudness, so volume="100%" doubles the gain rather than leaving it unchanged. Omit volume, or use medium, for no change. A gain that would clip is held under a limiter.
• Pitch keeps the speed. pitch="high" raises the voice without speeding it up, and speech marks keep their timing.
• Nested values multiply, as rate always has: a pitch="high" inside another pitch="high" is 1.15 times 1.15.
• A whole-block wrapper costs nothing extra. A prosody that changes part of a sentence has that part synthesized on its own, as rate already does, so the sentence’s intonation can break where the value changes.
• Billing is unchanged. A request is billed for its spoken text, and tags are not counted.

<emphasis> and <speechify:style emotion> are still accepted and not applied on these models. If you send pitch or volume today and do not want the change, remove the attribute.

No API version change is needed. See SSML.