TTS essentials

TTS as a Service (TTSaaS) is a text-to-speech Nuance engine that generates synthesized speech from text input.

TTSaaS receives input in form of plain text, SSML, or a combination of text and Nuance control codes. It returns synthesized speech as either an audio stream or a single audio package.

You choose the Nuance voice to render the speech from hundreds of voices available in different genders and many languages and locales. For the list of available voices in your region, see Geographies.

Voices

TTSaaS works with Nuance voices to render speech in many languages and locales, with different genders and styles available. Voices are available as standard and enhanced models.

You can list and filter voices programmatically to select the ones you want to use in your synthesis requests.

See Synthesizer API: GetVoicesRequest.

Input type

TTSaaS lets you input the text to synthesize in one of three formats:

  • Plain text, just one or more simple sentences.

  • SSML input consisting of text enclosed in SSML elements. TTSaaS supports most of the SSML elements defined in the WC3 SSML Specification Version 1.1.

  • Tokenized sequences, a combination of text and Nuance control codes. Most of these codes perform the same action as their SSML equivalents.

See Reference topics: Input to synthesize.

Audio formats

TTSaaS can generate speech in several audio formats and sampling rates. The default is PCM WAV audio at 22050 kHz but it also supports A-law, μ-law, Opus, and encapsulated Ogg (Ogg Opus).

See Synthesizer gRPC API > AudioParameters.

Resources

Apart from the mandatory voice resource, you can also include synthesis resources such as user dictionaries, ActivePrompt databases, rulesets, and audio files.

See Reference topics: Synthesis resources.

Synthesis and storage APIs

TTSaaS offers two synthesis APIs: a Synthesizer gRPC API for requesting synthesis, and a transcoded version as an HTTP API.

It also offers a Storage gRPC API for managing synthesis resources.

See:

Sample clients

You can experiment with TTSaaS using sample clients. Python clients for synthesis and storage are included in this documentation, along with instructions on how to use them.

See: