I’m creating a podcast and I need to convert my voice into a different voice while keeping my inton | ElevenLabs | Page 1

AI-generated Answer for joanamaciel_29235

Short answer: use ElevenLabs Voice Changer (formerly Speech‑to‑Speech). It converts one recorded voice into another while preserving your original timing, intonation, and delivery—ideal for anonymity without using TTS.

Two recommended workflows

No‑code (Creative Platform)

Optional cleanup: run your raw track through Voice Isolator to remove background noise before conversion. (help.elevenlabs.io)
In Voice Changer: upload an audio file or record directly in the browser, then pick any voice from your Voice Library (or your own cloned voice) and generate. Voice Changer is designed to preserve cadence/emotion and offers a built‑in “Remove background noise” toggle. (help.elevenlabs.io)
Constraints and cost: max 5 minutes per conversion (split longer episodes into segments); files under ~50 MB; priced at 1,000 credits per minute. (help.elevenlabs.io)
Supported input formats: audio (MP3, M4A, FLAC, OGA/OGG, WAV) and video (MKV, WEBM, MP4, MOV). (help.elevenlabs.io)
Exports: download WAV/M4A/FLAC from History after generation. (help.elevenlabs.io)

API/batch (automated post workflow)

Choose an output voice_id (e.g., from your Voice Library) and call POST /v1/speech-to-speech/:voice_id with your recorded audio. (elevenlabs.io)
Model: use the STS v2 models; eleven_multilingual_sts_v2 often performs best—even for English. (elevenlabs.io)
Useful parameters:
- remove_background_noise=true to suppress ambient sounds. (elevenlabs.io)
- enable_logging=false to use zero‑retention mode (disables history features). (elevenlabs.io)
- file_format=pcm_s16le_16 for low‑latency uncompressed input; seed and voice_settings for deterministic, consistent output. (elevenlabs.io)
Segment handling and quality: keep chunks ≤5 minutes; the model tracks emotion/cadence and retains language/accent from the source. (elevenlabs.io)
Pricing: billed by audio duration (1,000 credits/min). (help.elevenlabs.io)

Additional notes for podcasts

Voice selection: add public voices to “My Voices” from the Voice Library for an anonymized vocal identity. (help.elevenlabs.io)
Languages/accents: STS preserves cadence and accent; multilingual v2 supports 29 languages. (elevenlabs.io)
Optional transcripts for editing/redaction: use Speech‑to‑Text (Scribe v2). (help.elevenlabs.io)

Background material

Product intro and use cases for Speech‑to‑Speech/Voice Changer. (elevenlabs.io)

If you want, I can outline a concrete batch script example against

#I’m creating a podcast and I need to convert my voice into a different voice while keeping my inton