Google: Gemini 3.8 Flash TTS

Google: Gemini 3.8 Flash TTS is a text to speech model from Google, and it is the successor to Gemini 3.1 Flash TTS Preview. It sits in the creative tier of the 3.8 TTS family, which means Google positions it for expressive speech generation rather than for transcription or general text work. If you already route other Google models through api14, this one extends that same provider relationship into audio output.

The model takes text as its only input and produces speech as its only output. It has a context window of 32,768 tokens and a maximum output of 16,384, so you can send fairly long scripts in a single request and receive a correspondingly long spoken result. Thirty voices are available: Zephyr, Puck, Charon, Kore, Fenrir, Leda, Orus, Aoede, Callirrhoe, Autonoe, Enceladus, Iapetus, Umbriel, Algieba, Despina, Erinome, Algenib, Rasalgethi, Laomedeia, Achernar, Alnilam, Schedar, Gacrux, Pulcherrima, Achird, Zubenelgenubi, Vindemiatrix, Sadachbia, Sadaltager and Sulafat. That range lets you cast different speakers for different parts of a project without switching models.

Because the input is text and the output is speech, the natural fits are narration, voiceover, dialogue for games or video, audiobook style reading, and spoken versions of written content such as articles or documentation. The large context window helps when a single piece of content needs consistent delivery across many paragraphs, since you can pass the whole script rather than stitching together many short clips. The voice list also makes it practical to assign a distinct voice per character or per section.

You call it through api14's OpenAI compatible API using the model id "google/gemini-3.8-flash-tts". The request shape follows the same pattern as other models on the gateway, so existing OpenAI client libraries work without special handling. Note that this model lists no supported parameters, so plan on sending the model id plus your input text and voice selection rather than relying on tuning knobs like temperature or top_p. Check the live price table on the page for current rates before you commit to a workload.

One thing to weigh is scope. This model only accepts text and only returns speech, so it will not help with image, audio or document inputs, and it cannot produce written output. If your task is transcription, translation or general text generation, a text model is the better fit. Use Gemini 3.8 Flash TTS when the deliverable is audio and you want a choice of voices within a single request.