Synthèse Vocale Gratuite en Ligne - Texte en Voix MP3

The tool lets you download generated audio as MP3 (.mp3) and WAV (.wav). MP3 is encoded at a default 128 kbps CBR and 44.1 kHz sample rate, producing roughly 1 MB per minute; WAV exports are uncompressed 16-bit PCM at 44.1 kHz and produce about 10 MB per minute. After conversion you’ll see two buttons: “Download MP3” and “Download WAV”; the WAV button produces larger files suitable for editing, while MP3 is better for distribution and streaming.

Open the page, paste or type your text into the large “Enter text here” box (you can also upload a .txt file), select a voice from the “Select voice” dropdown, adjust Speed and Pitch sliders if needed, then click the blue “Convert” button. A progress bar appears; when finished you can click the in-page Play button to preview and then choose “Download MP3” or “Download WAV” to save the file. No account or email is required and the temporary audio file is available immediately after conversion.

Yes — the editor accepts up to 20,000 characters per conversion (roughly 3,000 words) and will convert up to about 15 minutes of generated audio in a single job. If you paste more than 20,000 characters the tool shows a red character counter and prompts you to split the text; large inputs on mobile may cause timeouts in Safari. If you need longer audio, split your document into sections, convert each, then stitch MP3s together in a desktop editor or an online audio joiner.

The dropdown labeled “Select voice” includes multiple neural-style voices such as English (US) Male/Female, English (UK) Female, Spanish (ES), French, and German; languages are shown by name next to the voice. Two sliders — Speed (50%–200%) and Pitch (-2 to +2 semitones) — let you change delivery, and a volume normalization checkbox ensures consistent RMS if you plan to combine clips. The interface does not currently support advanced SSML tags, so use the built-in pause/line-break controls to manage intonation.

WAV exports are uncompressed PCM 16-bit/44.1 kHz and preserve every sample, so they’re best for post-production or voiceover syncing; expect roughly 10 MB per minute. MP3 at the default 128 kbps reduces file size to about 1 MB per minute but introduces lossy compression artifacts that can slightly soften sibilance and breath sounds. Choose WAV if you’ll edit or remix in Audacity/Adobe Audition; use MP3 when you need small files for podcasts, mobile apps, or quick distribution.

The conversion runs on the server because the neural voice synthesis needs GPU-backed processing; the site uploads your text to a transient processing queue and returns a downloadable audio file. Generated audio files are kept temporarily and removed automatically after a short period; for immediate privacy concerns avoid pasting sensitive data. If you need an on-premise solution for regulated content, use a local TTS engine or desktop software instead.

Common causes are exceeding the 20,000-character limit, including unsupported emojis/special characters, or browser extensions blocking requests. First split the text under the character limit and remove odd Unicode characters (emojis often cause errors), then retry in Chrome or Firefox with extensions disabled; clear site data if you see persistent failures. If you hit a rate limit you’ll see a “Rate limit exceeded” message — wait 60 minutes or try again from a different network.

On desktop (Chrome/Firefox/Edge) conversions are stable and large previews stream without interruption; on mobile Safari and iOS Chrome long conversions can timeout or run out of memory, especially with texts near the 20,000-character limit. Mobile browsers may block autoplay of the preview — tap Play to allow audio — and iOS may prompt you to allow audio playback if not interacted with. For long documents, we recommend converting on desktop or splitting text into smaller chunks on mobile to avoid crashes.

The core converter is free to use with no signup and outputs audio files without watermarks; you can download MP3 and WAV files immediately after conversion. To prevent abuse there are soft rate limits (typically around 50 conversions per hour per IP); if you exceed that you’ll see a “Rate limit exceeded” notice and should wait about 60 minutes. There is no mandatory paid tier for basic use, but check the site footer for any announced premium features if you need batch processing or higher limits.

Audio files you download are provided for personal and commercial use under the site’s standard Terms of Use (see the Terms link in the footer for the full legal text). You should not use generated voices to impersonate private individuals, commit fraud, or violate rights — those are disallowed under the terms. For enterprise-scale licensing, trademarked voices, or voice cloning requests, contact the site’s support or consult the legal/terms page before large distribution.

Break very long paragraphs into shorter sentences and insert line breaks where you want natural pauses, because the engine interprets punctuation for prosody; avoid long lists without separators. Use the Speed slider in small increments (5–10%) rather than extremes, and choose WAV if you plan to apply post-processing like de-essing or EQ in Audacity/Adobe Audition. Finally, preview short segments before converting the whole chapter to confirm voice choice and pacing, which saves time given the conversion limits.

This online tool is a quick web-first generator for MP3/WAV export without installing software, but it doesn’t replace multitrack editing or advanced noise reduction available in Audacity/Adobe Audition. Export a WAV from the site (16-bit/44.1 kHz) and import it into Audacity or Audition for mixing, trimming, and compression; HandBrake is a video transcoder, so use it only when you need to mux the generated audio back into video. The web tool is best for fast voice generation, while desktop tools are better for detailed post-production and batch automation.