ElevenLabs Turbo v2.5 turns a script into speech with voice selection, adjustable pacing and controls for stability, similarity and style. Supply the words in the intended language and, for segmented narration, add preceding and following text as context. Text normalization guides how numbers and similar written forms are prepared for speech; optional timestamps return timing data with the audio. Prepare up to 40,000 Unicode characters after trimming, audition the voice settings and review the task’s character-based estimate before generating narration or spoken instructions.
ElevenLabs TTS Turbo v2.5 is developed by ElevenLabs. Vidgo provides access to the model. Model names and trademarks belong to their respective owners.Open the Turbo speech task, choose a voice name or compatible voice ID, and set the pace for the listener. Adjust stability, similarity and style while auditioning, then provide language, normalization and optional timing controls. Use previous_text and next_text when the submitted passage belongs to a larger narration, keeping the intended spoken segment in text.
The text field contains the words for this generated segment. previous_text and next_text provide the surrounding passage as optional context; voice and numeric controls define the speaking configuration. Use the intended spoken language in the source, then compare the actual audio with its script and settings. Page translations explain meaning while preserving the original recorded language.
| Model / task | Input and workflow |
|---|---|
| ElevenLabs TTS Turbo v2.5 Text to Speech | Turn scripts into multilingual speech with voice selection, adjustable pacing, vocal style controls, and context for segmented narration. |
The existing recordings include English origami instructions, Mandarin stargazing guidance with timestamps requested and a Spanish family memory with surrounding context. Each uses a distinct script and voice configuration. Listen to the results and inspect the source text to understand how the available controls support the intended delivery.
Use speed from 0.7 to 1.2 and write clear sentence boundaries for each action. The origami record uses Rachel at 1.05 with step-by-step directions; listen for word clarity and the time between instructions before choosing settings for your own guidance.
Supply the words in the intended language and set language_code when appropriate. The Mandarin stargazing request uses Aria, zh, speed 0.95 and timestamps true. Read all returned files to retrieve both the spoken audio and its separate timing or alignment attachment.
Add preceding and following text when a spoken segment belongs to a larger story. The Spanish memory record supplies context on returning through a doorway and opening the windows, while text contains the kitchen and two-cups-of-tea passage. Listen for how the segment fits its narrative tone.
Almost there. Fold the top corner down until it meets the center crease. Now press along the edge with your fingertip. That's it. Turn the paper over, and we'll make the next fold together.
Use complete sentences and punctuation to define the instructions or narrative progression. The displayed origami text is the actual English input. Review the delivered sequence against it and refine phrasing when the listener needs a clearer pause or transition.
{
"similarity_boost": 0.75,
"style": 0,
"apply_text_normalization": "auto",
"voice": "Sarah",
"language_code": "es",
"speed": 0.98,
"stability": 0.45,
"timestamps": false,
"previous_text": "Hacía años que no cruzaba aquella puerta.",
"next_text": "Después abrí las ventanas para dejar entrar la luz."
}Set stability, similarity_boost and style from 0 to 1, alongside voice, speed and normalization. The displayed Spanish configuration includes its actual surrounding text. Keep the complete configuration with the generated audio when comparing variations or integrating a narration workflow.
The origami request uses English instructions with Rachel and a slightly adjusted speed. The Spanish memory uses Sarah with previous and next passages to frame the submitted segment. Read the complete original texts, meaning translations and recorded voice settings while listening to their own results. Inspect pronunciation, pauses, pace and narrative tone in each recording.
Almost there. Fold the top corner down until it meets the center crease. Now press along the edge with your fingertip. That's it. Turn the paper over, and we'll make the next fold together.
{
"similarity_boost": 0.75,
"style": 0,
"apply_text_normalization": "auto",
"voice": "Rachel",
"language_code": "en",
"speed": 1.05,
"stability": 0.5,
"timestamps": false
}Original Spanish spoken text: Hoy volví a la casa de mi abuela. En la cocina seguía el reloj de siempre, cinco minutos adelantado. Preparé dos tazas de té por costumbre y, al darme cuenta, sonreí. Algunos gestos también guardan recuerdos. English meaning: Today I returned to my grandmother’s house. The same old clock was still in the kitchen, five minutes fast. Out of habit I prepared two cups of tea, and when I noticed, I smiled. Some gestures hold memories too. Original preceding context: Hacía años que no cruzaba aquella puerta. Meaning: It had been years since I last walked through that doorway. Original following context: Después abrí las ventanas para dejar entrar la luz. Meaning: Then I opened the windows to let in the light.
{
"similarity_boost": 0.75,
"style": 0,
"apply_text_normalization": "auto",
"voice": "Sarah",
"language_code": "es",
"speed": 0.98,
"stability": 0.45,
"timestamps": false,
"previous_text": "Hacía años que no cruzaba aquella puerta.",
"next_text": "Después abrí las ventanas para dejar entrar la luz."
}Prepare a script in the intended language, select the speaking voice and pacing, and add surrounding text when the segment is part of a larger narration. Generate and review the complete audio for spoken instructions, educational guidance, narrated stories, multilingual examples, longer segmented works and application speech. Request timestamps when you also need timing data.
Write each action as a clear sentence and set a suitable speaking speed. Listen to the generated sequence for word clarity and pauses before using it in a tutorial or guided activity.
Prepare a concept or observation in the learner’s language, choose voice and normalization, and request timing if needed. Review pronunciation and pacing for a lesson, museum explanation or science activity.
Write the spoken segment and supply preceding/following passages when they help establish its context. Audition voice character and pacing, then review continuity and emotional tone through the actual audio.
Prepare the exact words in each intended language and set its language code. Generate and review each recording’s pronunciation and rhythm before using it in a localized experience or learning exercise.
Divide the work into passages within the 40,000-character text limit and keep voice settings consistent. Supply the adjacent text as context, then listen across joins when assembling the reviewed recordings.
Collect text, voice and delivery settings, submit the documented request and retrieve audio by task ID. Read optional timestamp or alignment files as separate response entries for the application’s timing workflow.
Turbo v2.5 is billed proportionally at the current rate per 1,000 UTF-16 text units after leading and trailing whitespace is trimmed. Most English letters and Chinese characters count as one unit; some emoji and other supplementary characters count as two. The 40,000-character input limit is checked by Unicode characters, while the estimate uses the billing count. The dynamic table shows the current rate; enter the complete spoken text in the task to review the request amount and plan additional takes.
| Model | Provider | Workflow | Input material | Speech output | Starting rate |
|---|---|---|---|---|---|
| ElevenLabs TTS Turbo v2.5 Text to Speech | ElevenLabs | Text to Speech | Spoken text; optional context | Speech / optional timestamps | $0.04 / 1,000 billing units |
Test your creative brief online, then bring the verified request into your application.
Open the Turbo task and enter the full spoken segment, optional surrounding context and voice settings. Review the character-based estimate and generate. Listen for pronunciation, pace, pauses and narrative continuity, then retrieve all requested timing attachments with the audio before using the result.
Choose a task, inspect its settings and cost, and follow the result from an online test to an authenticated API request.
Prepare the exact segment to speak and choose a voice name or compatible voice ID. Select speed, stability, similarity and style, then language and normalization. Add previous_text and next_text when the segment belongs to a larger passage, and request timestamps when timing data is needed.
Open the Turbo task and enter the full spoken segment, optional surrounding context and voice settings. Review the character-based estimate and generate. Listen for pronunciation, pace, pauses and narrative continuity, then retrieve all requested timing attachments with the audio before using the result.
Use the complete spoken text and the task’s trimmed UTF-16 billing count with the current per-1,000-unit rate. Account for supplementary characters when reviewing the estimate and plan each revised take using its actual submitted text.
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.
ElevenLabs Turbo v2.5 is a Vidgo speech task with voice selection, speed, stability, similarity and style controls. It accepts up to 40,000 Unicode characters after trimming, supports optional preceding/following context, language and normalization inputs, and can return timing attachments with speech audio.
Read speech-model introductions, API guides and platform comparisons to understand script preparation, voice controls, integration and character-based costs.
