Elevenlabs Turbo V2.5 API Documentation
Playground
Try it on WaveSpeedAI!ElevenLabs Turbo V2.5 is a text-to-speech model available via WaveSpeedAI, billed at $0.05 per 1000 characters for TTS requests. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Features
Turbo V2.5 converts written text into natural, expressive speech with clear pronunciation, smooth pacing, and lively tone—ideal for voiceovers, narration, tutorials, podcasts, and digital content. It supports a rich library of multi-lingual voices and fast turnaround for production workflows. See the list here.
Key Features
- Fast, expressive synthesis with humanlike prosody
- Multilingual support plus strong English numeral/date reading
- Fine control via similarity and stability sliders
- Speaker Boost — enhances similarity to the original speaker’s voice.
Pricing
Billed by the exact character count of the input text, prorated — no rounding up to 1,000.
- Rate: $0.10 per 1,000 characters ($100 per 1M characters)
- No minimum charge; a 100-character request costs $0.010
| Characters | Cost |
|---|---|
| 100 | $0.010 |
| 500 | $0.05 |
| 1,000 | $0.10 |
| 10,000 | $1.00 |
How to Use
- Enter your script in the text field.
- Choose a voice_id (for example: Alicia, Darian, Eldrin). Refer to the full catalog in the voice list above.
- Optional controls • similarity: 0–1 (higher = closer to the base voice timbre) • stability: 0–1 (higher = more consistent delivery) • use_speaker_boost: enhances similarity to the original speaker’s voice
- Click Run to synthesize and preview your audio.
Notes
-
For steady rhythm, use clear punctuation and split very long text into shorter segments.
-
voice_id must be valid; if you see a voice-ID error, select one from the official list.
-
Speaker Boost may slightly increase generation latency.
-
Each request accepts 1–10,000 characters. Split longer text into separate requests.
-
Use
<break time="1.5s" />to add a pause of up to 3 seconds.
Authentication
For authentication details, please refer to the Authentication Guide.
API Endpoints
Submit Task & Query Result
set -euo pipefail
export WAVESPEED_API_KEY="your-api-key"
REQUEST_BODY=$(cat <<'JSON'
{
"text": "A clear example input",
"voice_id": "Alicia",
"similarity": 1,
"stability": 0.5,
"use_speaker_boost": true
}
JSON
)
# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
--connect-timeout 10 --max-time 60 \
-X POST "https://api.wavespeed.ai/api/v3/elevenlabs/turbo-v2.5" \
-H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
-H "Content-Type: application/json" \
-d "${REQUEST_BODY}"); then
printf 'Submit failed. %s\n' "${CHECK}" >&2
exit 1
fi
TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"
# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
# Transient errors (429, 5xx) are retried, honoring Retry-After.
HTTP_CODE=$(curl --silent --show-error --fail-with-body \
--connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
--output "${BODY_FILE}" --write-out '%{http_code}' \
"${RESULT_URL}" \
-H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
case "${HTTP_CODE}" in
2??) ;;
000|429|5??) # dropped connection or temporary error: keep polling
printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
sleep 10; continue ;;
*) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
cat "${BODY_FILE}" >&2; exit 1 ;;
esac
RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')
case "${STATUS}" in
completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
failed|cancelled|timeout|deleted)
printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
*) sleep 2 ;;
esac
doneRetry temporary errors only while polling, never the submit request. Errors, retries and timeouts
Parameters
Task Submission Parameters
Request Parameters
| Parameter | Type | Required | Default | Range | Description |
|---|---|---|---|---|---|
| text | string | Yes | - | 1 ~ 10000 characters | Text to convert to speech. Maximum 10,000 characters per request. Use <break time="1.5s" /> to add a pause of up to 3 seconds. |
| voice_id | string | Yes | Alicia | - | The voice to use for speech generation. Choose a preset voice name, or enter any ElevenLabs voice ID. |
| similarity | number | No | 1 | 0 ~ 1 | High enhancement boosts overall voice clarity and target speaker similarity. Very high values can cause artifacts, so adjusting this setting to find the optimal value is encouraged. |
| stability | number | No | 0.5 | 0 ~ 1 | Voice stability (0-1) Default value: 0.5 |
| use_speaker_boost | boolean | No | true | - | Enhances similarity to the original speaker's voice. May slightly increase generation latency. |
Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.
Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data.id | string | Unique identifier for the prediction, Task Id |
| data.model | string | Model ID used for the prediction |
| data.outputs | array | Output values, usually URL strings; some models return text strings or structured result objects (empty when status is not completed) |
| data.urls | object | Object containing related API endpoints |
| data.status | string | Task status. completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses. |
| data.created_at | string | ISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”) |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |
Result Request Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
| id | string | Yes | - | Task ID |
Result Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data | object | The prediction data object containing all details |
| data.id | string | Unique identifier for the prediction |
| data.model | string | Model ID used for the prediction |
| data.outputs | array<string | object> | Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model. |
| data.urls | object | Object containing related API endpoints |
| data.status | string | Status: completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses |
| data.created_at | string | ISO timestamp of when the request was created |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |