Fish Audio

Fish Audio is a voice AI platform for text-to-speech, voice cloning, speech-to-text, voice changing, and conversational voice applications. Its web studio lets creators type scripts, select voices, control emotion and special tags, and generate speech; its APIs support low-latency text-to-speech, transcription, voice cloning, and end-to-end voice agents. Fish Audio also offers Story Studio, audio separation/translation, sound effects, a public voice library, and multilingual generation.

✨ Key Features

• Generate expressive text-to-speech — Turn scripts into natural narration with controllable tone and emotion.
• Control emotion with tags — Add tags such as emphasis, whispering, excitement, laughter, crying, and pauses.
• Clone voices — Create a digital voice from a short reference recording and use it across supported workflows.
• Use a large voice library — Browse more than 2 million user-uploaded voices for creative and production starting points.
• Transcribe speech — Use speech-to-text with multispeaker support and natural-language descriptions.
• Build voice agents — Connect low-latency voice capabilities to conversational assistants and customer support.
• Generate video voiceovers — Match narration to videos, ads, explainers, and other scripted content.
• Produce audiobooks — Create long-form narration with pacing, emotion, and chapter-level control.
• Create character voices — Craft voices and personas for games, animation, and interactive stories.
• Support multilingual speech — Generate speech in 30+ languages with supported voices.
• Use developer APIs — Integrate TTS, ASR, cloning, streaming, and voice-agent capabilities into apps.
• Explore audio tools — Use voice changing, Story Studio, audio separation, audio translation, and sound effects where supported.

💳 Pricing

• Free — $0 with 8,000 credits/month for personal, non-commercial generation. Commercial use and monetization require a paid plan according to the official FAQ.
• Plus — $15/month monthly or $132/year annually; 250,000 monthly credits, up to 200 minutes of generation, up to 15,000 characters per generation, public/private voice slots, priority generation, Voice Design, professional voice space, enhanced cloning, and commercial use.
• Pro — $100/month monthly or $900/year annually; 2,000,000 monthly credits, up to 1,620 minutes of generation, 3 team seats, up to 30,000 characters per generation, unlimited voice slots, 5 professional voice spaces, and a 7-day refund guarantee.
• Max — $999/month paid monthly or $8,988/year (equivalent to $749/month), with 25,000,000 credits/month, up to 6,250 generation minutes, and 10 team seats on the pricing page checked October 1, 2026.
• API usage — Pay-as-you-go: TTS models s2-pro and s1 at $15 per million UTF-8 bytes; ASR transcribe-1 at $0.36 per audio hour. Enterprise volume pricing and rate limits are custom.

The free plan is for personal use, while paid plans provide commercial rights. Check the current plan page for regional pricing, credits, and model availability.

⚖️ Alternatives Compared

🎯 Use Cases

• Creating multilingual YouTube or course voiceovers.
• Narrating audiobooks without a recording booth.
• Designing recurring character voices for games or animation.
• Building a customer-support chatbot with a natural voice.
• Cloning an approved voice for repeatable content production.
• Transcribing interviews or multi-speaker recordings.
• Adding emotional delivery and pauses to scripted narration.
• Integrating low-latency TTS or ASR into an application.
• Producing ads, explainers, podcasts, and social-video narration.
• Separating, translating, or transforming audio in a creative workflow.