Chatterbox Speech To Speech

Chatterbox Speech To Speech

Playground

Chatterbox Speech to Speech is a fast AI voice conversion model that converts source audio into a target voice style with optional reference audio guidance. Ready-to-use REST inference API for voice conversion, speech style transfer, dubbing, character voices, creator content, audio localization, and professional speech-to-speech workflows with simple integration, no coldstarts, and affordable pricing.

Features

Chatterbox Speech-to-Speech transforms a source audio clip into a target voice style using optional reference audio. It is suitable for voice conversion, style transfer, creator dubbing, character voice prototyping, and other speech-to-speech workflows where you want to preserve spoken content while changing vocal identity or delivery style.

Why Choose This?

• Speech-to-speech conversion
  Transform an existing speech recording into a different voice style.

• Optional reference voice guidance
  Add reference_audio when you want the output to follow a particular vocal tone or character.

• Simple workflow
  Upload source audio, optionally upload a reference voice sample, and generate the converted result.

• Useful for creator and dubbing workflows
  Suitable for voice restyling, character voice tests, demo production, and spoken-content transformation.

• Production-ready API
  Useful for narration replacement, voice experiments, content localization, and creative audio workflows.

Parameters

How to Use

1. Upload your source audio — provide the speech recording you want to transform.
2. Upload reference audio (optional) — add a target voice sample if you want stronger style guidance.
3. Submit — run the model and download the converted speech audio.

Example Use Case

Convert a spoken voice clip into a different vocal style for creator content, dubbing, or character voice testing.

Pricing

Just $0.02 per started minute.

Billing Rules

• Pricing is $0.02 per started minute
• Audio duration is billed in started 60-second units
• Audio shorter than 60 seconds is billed as 1 minute
• reference_audio does not affect pricing

Example Costs

Best Use Cases

• Voice style transfer — Convert speech into a different vocal tone or identity.
• Character voice prototyping — Test alternative voice styles for characters or avatars.
• Creator dubbing — Rework spoken audio for short-form content or promos.
• Narration restyling — Preserve content while changing delivery feel.
• Speech workflow experiments — Compare different voice directions from the same recording.

Pro Tips

• Use clean source audio for better intelligibility.
• Add reference_audio only when you want stronger target voice guidance.
• Use a clear reference sample with stable tone for more consistent conversion.
• Short clips are useful for testing before processing longer audio.

Notes

• audio is required.
• reference_audio is optional.
• Pricing is based on source audio duration and billed per started minute.
• Better source audio and cleaner reference audio generally improve output quality.

Related Models

• Chatterbox Text-to-Speech — Generate speech directly from text.
• Voice cloning workflows — Useful when you need a reusable custom voice identity instead of per-request voice guidance.
• Audio generation workflows — Useful when you need music or sound generation instead of speech conversion.

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'
{
  "audio": "https://interactive-examples.mdn.mozilla.net/media/cc0-audio/t-rex-roar.mp3"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/chatterbox/speech-to-speech" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters