Hearing a draft can reveal an awkward transition or a missing word that your eyes keep skipping. Text to Speech sends entered text to the browser’s speech-synthesis API using voices available on your device. The voice list and playback behavior depend on the browser and operating system.
Choose a voice and pace
Paste up to 5,000 characters. Choose System default or a named voice from the menu, then set Speaking rate between 0.5 and 2. Start at 1 to hear the voice’s normal rate. A slower setting can make proofreading easier, but the exact pacing varies between voices.
Select Read aloud. Use Stop reading when you have heard enough or want to change a sentence. Starting a new reading cancels the previous speech request. Work in short sections when reviewing a long draft, so you can make corrections without losing your place.
A voice marked local is reported by the browser as a local service. Other voices may depend on a network service. ERITHRA does not guarantee that every voice is installed, available offline, or equally capable in every language.
Compare one sentence at two rates
Enter The next stop is Central. Choose an English voice and rate 1, then read it aloud. Stop playback, set the rate to 0.75, and read the same text again. The second request asks the voice to speak more slowly; it does not change the sentence or generate a separate audio file.
Now add a comma or replace an ambiguous word and listen again. Punctuation can help you assess rhythm, although different voices may interpret it differently. Treat unusual pronunciation of a name as a voice limitation before deciding that the written name is incorrect.
The returned text remains available for review, but there is no MP3 or WAV export. Browser speech playback is the output of this tool. Recording or generating a distributable narration requires a different workflow.
Handle unavailable speech gracefully
If speech synthesis is unsupported, the tool reports that limitation. If a particular voice fails, try System default or another listed voice, and check device audio settings. Voice availability can change after the browser finishes loading its voice list.
Entered text is processed through native browser speech facilities; network use depends on the selected service. Consider that before reading confidential material with a nonlocal voice. Use Word Counter to divide a long draft into manageable sections. Speech to Text provides the opposite direction for dictation, with different browser support and privacy considerations.