AI Voice Studio: Text to Speech, Voice Cloning and Transcription

AI Voice Studio is a website with three voice tools on one page: Text to Speech, Voice Clone and Speech to Text. You can type text and get an MP3 in one of 23 Standard or 16 HD preset voices, have the text spoken in a voice cloned from a short recording, or upload audio or video and get a transcript with SRT subtitles. You use it in your browser while the audio is generated in the cloud, you sign in with Google, and all three tools use the same credit balance.

You can start for free: signing in with Google for the first time gives you 20 credits, valid for 365 days. That’s enough for 4,000 characters of Standard speech or voice cloning, about 660 characters of HD speech, or 20 minutes of transcription. Daily check-ins add more: 2, 2, 4, 2, 2, 2 and 6 credits over a 7-day streak, 20 in total, and those credits last 30 days. Missing a day restarts the streak at day 1. For more, there are monthly plans, whose credits are valid until the end of each billing month, and one-time credit packs, whose credits never expire.

Open the Voice Clone tab, upload or record a sample, tick the consent box, type your text and press Generate. The sample needs at least 3 seconds of speech and only the first 20 seconds are used; 5–15 seconds of one person talking in a quiet room works best. Files can be MP3, WAV, M4A, WebM or OGG, up to 10 MB. Each request takes up to 1,000 characters and costs 1 credit per 200 characters. Only clone your own voice or a voice you have permission to use. Our voice cloning guide walks through each step.

Voice Clone works in 21 languages: Chinese, English, German, Spanish, French, Japanese, Italian, Hungarian, Korean, Russian, Persian, Arabic, Polish, Portuguese, Dutch, Czech, Danish, Swedish, Greek, Turkish and Hebrew. Pick the language your text is written in from the dropdown. If your sample is in a different language from your text, run a short test first, since results vary. Hindi and other languages outside this list can’t be cloned, but the HD preset voices read 29 languages, Hindi included.

Standard costs less and HD covers more languages. Standard has 23 voices that each speak English, French, German, Spanish, Italian, Portuguese and Dutch; it costs 1 credit per 200 characters, takes up to 3,000 characters per request and runs on our own servers on an open-source model. HD has 16 premium voices from a commercial cloud voice model, detects the language of your text across 29 languages, lets you set the speed from 0.7× to 1.2×, and costs 3 credits per 100 characters with up to 5,000 characters per request. A 1,000-character script costs 5 credits in Standard and 30 in HD. HD text is sent to a third-party voice provider to generate. Play the previews to compare.

Yes. Upload an MP3, WAV, M4A, MP4, MOV, WebM, OGG or FLAC file up to 30 MB and 15 minutes long, and Speech to Text returns the full transcript. The language is detected automatically, or you can set it yourself; about 99 languages are supported. Copy the text, or download it as a TXT file or as SRT subtitles with timestamps for your video editor. It costs 1 credit per minute of audio, rounded up, so a 15-minute file costs 15 credits.

Your voice sample is processed on our own servers and is not sent to any third-party AI company, and we don’t use your uploads or your text to train models. Voice Clone, Standard speech and transcription all run on our servers with open-source models. The one exception is HD speech: the text you type is sent to a third-party voice provider to generate the audio. Your uploads and results are kept in cloud storage for your account for 30 days, then deleted automatically; after you’ve made a purchase, new uploads and results are kept for 6 months. Download anything you want to keep, and email [email protected] if you’d like your files deleted sooner.

Yes. You can use audio made with Standard voices, HD voices and Voice Clone commercially, for example in monetized YouTube or TikTok videos, ads, client work, paid courses and podcasts. No watermark is added to any audio. With Voice Clone the usual rule still applies: only clone your own voice or a voice whose owner has agreed.

Yes. If you’re not happy with a purchase, email [email protected] and we’ll refund the payment to your original card or payment method, as money rather than credits. Separately, if a generation or transcription fails, the credits it used go back to your balance automatically. You can also cancel a plan anytime in Settings → Billing: you won’t be charged again, the plan stays active until the end of the period you’ve already paid for, and credits already added can be used until they expire.

No. AI Voice Studio runs in your web browser, so there’s nothing to download and you don’t need a graphics card: all generation happens on our servers. Sign in with your Google account and the studio is ready, and you can record a voice sample straight from the browser with your microphone. There’s no desktop or mobile app to install.