ElevenLabs Text to Speech WAV: Format & Download

Yes. ElevenLabs text to speech can output WAV files through the output_format API parameter, which includes WAV alongside MP3, PCM and other formats. The default is MP3 at 44.1kHz, and WAV at 44.1kHz requires a Pro tier subscription or above. Plans start at $6/month for Starter with 30,000 credits (checked 5 Nov 2026).

WAV suits finished masters; the streaming guide covers live audio instead, the Multilingual V2 guide helps you pick a model, the troubleshooting guide fixes failed generations, and the Dubbing V2 guide covers video audio.

How WAV output works on ElevenLabs

The text to speech API accepts an output_format query parameter that controls the audio encoding. The documented default is mp3_44100_128 – MP3 at 44.1kHz and 128kbps – which suits general use. To get WAV, you pass a WAV value for output_format instead. The API reference notes two tier gates: PCM and WAV formats with 44.1kHz sample rate require a Pro subscription or above, and MP3 with 192kbps bitrate requires a Creator subscription or above. In other words, high-end output formats are a paid-tier feature, and the very best quality is reserved for the top plan.

WAV is uncompressed, which means larger files but zero compression artifacts. A minute of 44.1kHz 16-bit stereo WAV is about 10MB, versus about 1MB for the same minute as 128kbps MP3. That tradeoff matters when the audio will be processed further: every MP3 encode-decode cycle loses a little quality, while WAV survives unlimited editing. For final delivery to listeners, MP3 or Opus is usually the better choice; for production work, WAV is the professional standard that studios expect.

The 44.1kHz sample rate is the CD audio standard, which is why ElevenLabs gates it behind Pro. Lower sample rates like 8kHz or 16kHz are sufficient for telephony and voice assistants, but music production and broadcast expect 44.1kHz or higher. If your workflow involves a digital audio workstation, the WAV container with 44.1kHz PCM is the most compatible choice – it imports cleanly into Pro Tools, Logic, Ableton, Audacity and every other major editor without conversion.

How to get WAV audio, step by step

WAV output is available through the API, not as a button in the basic web player (which downloads MP3). These steps assume you have an API key and a Pro-tier subscription for 44.1kHz WAV.

1. Confirm your plan tier. WAV at 44.1kHz requires Pro ($99/month, 600,000 credits) or above. Lower sample-rate PCM options may be available on lower tiers – check the output_format enum in the docs. Cost: the plan itself; no per-request surcharge for WAV.
2. Get an API key. Create one in your ElevenLabs profile. The xi-api-key header authenticates the request. Cost: free to create.
3. Call the convert endpoint with output_format. POST to the text-to-speech endpoint with your voice id and text, and set the output_format query parameter to a WAV value. Our Python quickstart shows the request structure. Cost: billed per character at the model rate – about 1 credit per character on standard models.
4. Save the response as .wav. The API returns raw audio bytes in the requested format. Write them to a file with a .wav extension. A 1,000-character generation costs about 1,000 credits and produces roughly a minute of WAV audio.
5. Verify the file. Open it in your editor or player and confirm the sample rate and bit depth match what you requested before batching a large job. Cost: free to inspect; re-generating costs credits again.

Available output formats compared

The output_format enum covers the practical range from compact to lossless. MP3 options span bitrates for different quality needs, PCM gives raw samples at several sample rates, and WAV packages PCM in the standard container. Opus serves low-bitrate streaming. Your tier gates the top end: Creator includes 192kbps MP3, Pro includes 44.1kHz PCM and WAV.

When WAV is worth it

Choose WAV when the audio will be edited, mixed or mastered. Music producers layering AI vocals, podcast engineers doing noise reduction and EQ, and video editors syncing dialogue all benefit from starting with lossless source. WAV also matters for archival: a WAV master can be transcoded to any future format without generational loss, while an MP3 master bakes in its compression forever. If you are building a sound library or sample pack, WAV is the only acceptable source format.

Skip WAV when the file goes straight to listeners. Streaming platforms recompress uploads anyway, web players handle MP3 natively, and mobile users on limited data prefer smaller files. A good rule: produce in WAV, deliver in MP3 or Opus. The API makes this easy – generate once in WAV for your archive, then transcode copies for each delivery target. This two-step workflow gives you the best of both worlds: pristine masters and efficient distribution.

What WAV output costs

The format does not change the credit cost – you pay per character generated, not per megabyte. A 5,000-character script costs about 5,000 credits on a standard model whether you request MP3 or WAV. What changes is the plan you need: 44.1kHz WAV requires Pro at $99/month with 600,000 credits. If you only need WAV occasionally, generate on Pro for one month, download your masters, then drop back to a cheaper tier. The 600,000 credits on Pro generate roughly 600,000 characters of audio, which is about 100 hours of WAV masters.

Pros

• True lossless output with zero compression artifacts for production work.
• Same per-character credit cost as MP3, with no surcharge for WAV.
• Standard container opens in every audio editor and DAW.
• Ideal archival master that can be transcoded to any future format.
• PCM options at multiple sample rates for specialized pipelines.

Cons

• 44.1kHz WAV requires Pro tier at $99/month or above.
• API-only, since the web interface downloads MP3 and not WAV.
• Files are roughly 10x larger than 128kbps MP3 for the same audio.
• Overkill for direct-to-listener delivery where MP3 suffices.

Who should NOT bother with WAV

Do not pay for Pro just to get WAV if your audio goes straight to YouTube, Spotify or a web player – those platforms compress everything on upload, so your pristine WAV becomes their bitrate anyway. Do not use WAV for voice agents or IVR either; the larger chunks add network latency for no audible benefit in a phone call. And if you are on Free, Starter or Creator, do not waste time hunting for a WAV button that is not there – the web UI exports MP3, and 44.1kHz WAV is documented as a Pro-and-above API feature.

Frequently asked questions

Can ElevenLabs generate WAV files?

Yes, through the API. The output_format parameter accepts WAV values, and the docs confirm WAV is a supported output format. The web interface downloads MP3; for WAV, call the text-to-speech endpoint with output_format set accordingly.

What plan do I need for WAV at 44.1kHz?

Pro or above. The API reference states that PCM and WAV formats with 44.1kHz sample rate require a Pro subscription or above. Pro is $99/month with 600,000 credits (checked 5 Nov 2026). Lower tiers can use the default MP3 output.

Does WAV cost more credits than MP3?

No. Credits are charged per character generated, not by output format or file size. A 1,000-character generation costs about 1,000 credits on a standard model whether you request MP3, WAV or PCM. The format only affects which plan tier you need.

What is the default output format?

The documented default is mp3_44100_128 – MP3 at 44.1kHz sample rate and 128kbps bitrate. This applies when you do not specify output_format. It is a sensible default for general use: compact files with quality fine for most listening.

Can I get WAV from the ElevenLabs website?

Not directly. The web interface and history downloads provide MP3. WAV output is an API feature via the output_format parameter. If you need WAV without coding, generate the MP3 and convert it – but note that converting MP3 to WAV does not restore quality lost in the original MP3 compression.

Is WAV better than MP3 for voiceovers?

For production, yes – WAV is lossless, so it survives editing, EQ and mixing without generational loss. For delivery to listeners, the difference is inaudible in most cases and MP3 files are about 10x smaller. Produce in WAV, deliver in MP3 is the professional workflow.

What sample rates does the API support?

The output_format enum includes PCM at 8kHz, 16kHz, 22.05kHz, 24kHz and 44.1kHz, plus WAV and MP3 variants. The 44.1kHz PCM and WAV options require Pro or above. Lower sample rates suit telephony (8kHz) while 44.1kHz is the CD standard for music and broadcast.

How large is a WAV file from ElevenLabs?

About 10MB per minute for 44.1kHz 16-bit stereo, versus about 1MB per minute for 128kbps MP3. A 10,000-character generation (roughly 10 minutes of audio) produces about 100MB as WAV. Plan storage accordingly when batching WAV output.

Can I convert ElevenLabs MP3 to WAV myself?

Yes, with any audio converter or editor, but the result is not true lossless – it is the MP3 audio repackaged in a WAV container, with the original compression artifacts intact. For genuine lossless quality, generate directly in WAV through the API on a Pro-tier plan.

Sources checked

• elevenlabs.io/docs/api-reference/text-to-speech – output_format parameter, WAV support, Pro tier requirement for 44.1kHz WAV, checked 5 Nov 2026.
• elevenlabs.io/pricing – plan prices and monthly credit allowances, checked 5 Nov 2026.