Voice cloning
Fish Speech is built for voice cloning, which means speech that copies a voice.
Turn text into speech
High-quality voice cloning
Opens the audio studio with Fish Speech ready to use. You will need a free account to run it.
What it makes
Turn text into speech
Reads your text aloud in a natural-sounding voice.
What you need
Optional: the audio to work on.
What you get
A spoken audio clip of your text
Fish Speech, from Fish Audio, is a voice cloning model. Voice cloning means making new speech that copies a voice. You write your text in the Prompt field and get spoken audio back. You can also add one audio clip.
Prompt is the only required input. The audio clip is optional, and Fish Speech accepts one clip at a time. Its languages are Arabic, German, English, Spanish, French, Italian, Japanese, Korean, Dutch, Polish, Portuguese and Russian.
Fish Speech has no other options. It costs 200 credits.
Same job: turn text into speech. Compare before you choose.
CSM (Conversational Speech Model) is a speech generation model from Sesame that generates RVQ audio codes from text and audio inputs.
A fast and expressive Hindi text-to-speech model with clear pronunciation and accurate intonation.
A highly efficient Mandarin Chinese text-to-speech model that captures natural tones and prosody.
A fast and natural-sounding Japanese text-to-speech model optimized for smooth pronunciation.
Fish Speech is built for voice cloning, which means speech that copies a voice.
Arabic, German, English, Spanish, French, Italian, Japanese, Korean, Dutch, Polish, Portuguese and Russian.
Fish Speech accepts one audio clip with your Prompt.
Type the text you want spoken in the Prompt field.
You can attach one audio clip. Leave it out to run with the prompt alone.
Run Fish Speech to get spoken audio back.
Turn a written script into spoken audio.
Make a spoken track from your text in any of its twelve languages.
Quick answers to help you get the most out of Fish Speech in FunAI Studio.