Gemini 3.1 Flash TTS turns a script into multilingual narration or two-speaker dialogue. Select from 30 voices, describe the desired delivery in style_instructions and place audio tags where emotion or pacing should change. For dialogue, assign a voice to each of two speaker aliases and label the corresponding turns in the text. Choose language, temperature and MP3, WAV or Ogg Opus output for the intended workflow, then audition the complete recording. The estimate includes the Unicode characters in both the script and style instructions.
Gemini 3.1 Flash TTS is developed by Google. Vidgo provides access to the model. Model names and trademarks belong to their respective owners.Use voice for a narrator, or provide exactly two speakers with distinct aliases and assigned voices for dialogue. Put the words and inline cues in text and the overall performance direction in style_instructions. Open the existing task to choose output settings and review the full character estimate before generating.
The text field holds spoken words, audio tags and dialogue labels. style_instructions supplies the overall tone, pace or accent guidance. In two-speaker mode, use speaker_id values made from letters, digits or underscores and prefix each turn with the matching alias and a colon. Select a supported language value or let the model detect the language from the script.
| Model / task | Input and workflow |
|---|---|
| Gemini 3.1 Flash TTS Text to Speech | Create multilingual narration and two-speaker dialogue with 30 voices, style instructions and expressive audio tags. |
The recorded examples include an English rehearsal with two voices, a Mandarin photography guide and British-English cycling commentary. Each has its own script, style direction and voice settings. Listen to these separate recordings and inspect the actual inputs to understand how speaker assignment, delivery cues and output configuration are used.
Provide exactly two distinct speaker aliases and select a supported voice for each. The rehearsal record assigns Puck to Actor and Kore to Director, then labels their alternating lines. Listen to the role changes and delivery within the complete scene.
Use a single voice and style_instructions for tone, pace, emphasis or accent guidance. The Mandarin gallery record asks for warm, natural speech and gentle pauses with Sulafat, giving you a concrete source for reviewing narration against the photograph description.
Place appropriate inline tags near the words whose pace or emotion should change and support them with a clear style brief. The cycling record moves from focused commentary to a sprint and a warm finish; inspect clarity and transitions in the returned audio.
Actor: [whispers] You're early. The curtain isn't up yet. Director: Same words again. This time, you've been waiting years to see them. Actor: [excitement] You're early! The curtain isn't up yet! Director: [laughs] There it is. Keep that smile. We're ready.
Prefix each line with its assigned speaker alias and describe the delivery where it changes. The displayed rehearsal text is the original English input, including Actor/Director labels and whispered or excited cues. Compare the spoken turns with that complete script.
{
"style_instructions": "A warm Mandarin photography gallery guide. Speak naturally with gentle pauses. Softly emphasize the light in the window. Read only the Chinese narration, without speaking the bracketed directions.",
"voice": "Sulafat",
"language_code": "Chinese Mandarin (China)",
"temperature": 1,
"output_format": "mp3"
}Choose a supported language value, temperature from 0 to 2 and output_format mp3, wav or ogg_opus. The shown gallery settings preserve its actual Mandarin language value, Sulafat voice and MP3 output. Select the configuration for playback or editing and audition the result.
The rehearsal request gives Actor and Director separate voices, then asks the Actor to repeat a line with a different emotional intent. The gallery request supplies Mandarin narration about reflected street light and a warm window, with a gentle style brief. Read the original scripts, meaning translations and full settings while listening to each result, including voice changes, pauses and emphasis.
Actor: [whispers] You're early. The curtain isn't up yet. Director: Same words again. This time, you've been waiting years to see them. Actor: [excitement] You're early! The curtain isn't up yet! Director: [laughs] There it is. Keep that smile. We're ready.
{
"style_instructions": "An intimate theatre rehearsal. Actor shifts from a secretive whisper to delighted recognition. Director is calm, warm and lightly amused. Keep both voices distinct; do not read speaker labels or tags aloud.",
"speakers": [
{
"speaker_id": "Actor",
"voice": "Puck"
},
{
"speaker_id": "Director",
"voice": "Kore"
}
],
"language_code": "English (US)",
"temperature": 1,
"output_format": "mp3"
}Original Mandarin spoken text: 请看照片右侧的雨伞。[short pause] 伞面上的水珠，映着街边的灯光。再看远处那扇窗。[slow] 一点暖光，照亮了安静的小巷。摄影把匆匆经过的日常，留成了值得细看的风景。 English meaning: Look at the umbrella on the right side of the photograph. [short pause] The droplets on its surface reflect the lights beside the street. Now look at that window in the distance. [slow] A little warmth illuminates the quiet alley. Photography preserves the everyday moments we hurry past as scenes worth looking at closely.
{
"style_instructions": "A warm Mandarin photography gallery guide. Speak naturally with gentle pauses. Softly emphasize the light in the window. Read only the Chinese narration, without speaking the bracketed directions.",
"voice": "Sulafat",
"language_code": "Chinese Mandarin (China)",
"temperature": 1,
"output_format": "mp3"
}Prepare a narrator’s script or a two-role conversation, assign the voices and describe the intended performance. Generate in the selected format and listen through the full result before using it for story dialogue, gallery guidance, educational explanations, event commentary, localized narration or speech in an application.
Write alternating lines for two named roles and assign a voice to each. Describe their relationship and delivery in the style instructions, then review turn changes and emotional progression in the generated scene.
Prepare a description of the object or image and indicate where attention should move. Select a voice and a clear delivery style, then review pacing and emphasis for an audio guide.
Write an explanation with clear examples and planned pauses, then describe the intended speaking style. Listen for terminology, pronunciation and continuity before using the audio in a lesson.
Describe the event progression and use delivery cues where energy changes. Review the complete speech for clear words, rising intensity and the final reaction while developing a commentary concept.
Prepare the actual script in the target language and select a supported language value and voice. Audition pronunciation, rhythm and style in each generated version before placing it in the localized experience.
Collect script, style direction and voice or two-speaker assignments, then submit the documented request. Retrieve the finished audio by task ID and preserve its chosen format for playback or further editing.
Gemini TTS uses the current rate per 1,000 Unicode characters for text plus style_instructions. The existing task counts the supplied strings, including their spaces and inline delivery tags, and calculates the amount proportionally. The dynamic table shows the current character rate; enter both complete fields in the task to review the full estimate before submitting and plan revised takes from their actual input lengths.
| Model | Provider | Workflow | Input material | Audio formats | Starting rate |
|---|---|---|---|---|---|
| Gemini 3.1 Flash TTS Text to Speech | Text to Speech | Text; one or two speakers | MP3 / WAV / OGG_OPUS | $0.12 / 1,000 input characters |
Test your creative brief online, then bring the verified request into your application.
Open the Gemini TTS task and enter the complete text and style instructions. Configure the narrator or two speakers, language, temperature and output format, then review the combined character estimate. Generate and listen through all spoken turns for pronunciation, emotion, pauses and role changes before using the audio.
Choose a task, inspect its settings and cost, and follow the result from an online test to an authenticated API request.
Prepare a single narrator or exactly two dialogue roles. Choose a voice for the narrator, or distinct speaker aliases with an assigned voice for each role and matching labels in the script. Add style direction and optional inline cues, then choose language, temperature and output format.
Open the Gemini TTS task and enter the complete text and style instructions. Configure the narrator or two speakers, language, temperature and output format, then review the combined character estimate. Generate and listen through all spoken turns for pronunciation, emotion, pauses and role changes before using the audio.
Count the Unicode characters in both the script and style instructions, including supplied spaces and delivery tags. Use the task’s current per-1,000-character rate and review the combined estimate before each take.
Read the original spoken text and voice settings while listening to the result. Translations explain the meaning while preserving the recording’s original spoken language. Review pronunciation, pacing and expression before adapting the request to your own script.
Retain the task ID returned by the API to query status and retrieve completed file URLs. Use your online generation history to review earlier results and the creative direction you tested.
Use the documentation linked from each task for its exact model identifier, accepted inputs, authentication and response format. Integrate the same task you tested online so its media and settings match the request.
Gemini 3.1 Flash TTS is a Google speech model on Vidgo for narration and exactly two-speaker dialogue. It provides 30 voices, multilingual input, style instructions, inline audio tags, temperature control and MP3/WAV/Ogg Opus output. The text field accepts up to 50,000 Unicode characters.
Read speech-model introductions, API guides and platform comparisons to understand script preparation, voice controls, integration and character-based costs.
