Cycling through all six. Tap any point to stop.

Measured on six things

AI text-to-speech platform with director-mode emotion control, 30-second voice cloning and custom voice design. Free to start, paid from $6/mo.

Speed

Lighthouse mobile performance 55/100 with LCP 20.8s and FCP 18.8s — very slow initial load on mobile; desktop better at 63/100.

Screenshots

1 / 1

Soundwaver

About this app

Soundwaver is an AI text-to-speech platform built on the MiMo-V2.5-TTS model, designed for Chinese-language content creators. Director Mode lets you control delivery with plain-language instructions (e.g. "excited, slightly fast, with a smile") instead of tweaking endless sliders. Clone a voice from a 30-second sample, or design a brand-new voice with WaveMind simply by describing it in words. A community voice library lets creators discover and use voices shared by others. The free plan includes 1,000 credits (approx. 50,000 Chinese characters) with 1 voice slot and 1 clone slot; paid plans start at $6/mo (7,200 credits) with up to 4,000 characters per generation — chapter-length text for audiobooks. Ideal for short-video narration, audiobooks and e-learning. Bilingual Chinese/English interface, secure checkout via Stripe.

Key features

1Director Mode: control delivery with plain-language instructions (role, scene, direction) instead of sliders.
2Voice cloning from 30-second sample; WaveMind voice design from text description.
3Community voice library; bilingual Chinese/English interface; free plan with 1,000 credits.

Target customer profile

Drawn from the product itself, not from a survey.

Demographic

Content creators (YouTubers, short-video makers) aged 18-45, tech-savvy, active on social media

Pain points

Need engaging voiceovers but lack recording equipment or voice-over skills; time-consuming to produce; budget constraints for professional voice actors

Primary needs

Quick, natural-sounding voice generation; easy workflow; affordable pricing; ability to clone own voice for consistent branding

Demographic

Online educators and course creators in Taiwan/Hong Kong or teaching Traditional Chinese

Pain points

Large volume of lecture materials need narration; keeping content up-to-date requires frequent re-recordings; may have non-native accents

Primary needs

Reliable and natural TTS in Traditional Chinese; ability to edit scripts without re-recording entirely; high-quality audio for professional courses; tools to clone their own voice for personal touch

Demographic

Podcasters and audiobook narrators (amateur to professional) focusing on Mandarin/Taiwanese content

Pain points

High production costs for audiobook narration or podcast segments; rigid schedules for studio sessions; need to produce multi-episode content rapidly

Primary needs

High-quality AI voices that maintain emotional nuance; batch processing for series; convenient script-to-speech with fine-tuned control over pacing and tone; commercial rights for paid plans

Demographic

Marketing professionals and business owners in Taiwan or targeting Tang-Min community

Pain points

Need localized ads or product demos without recording studios; limited internal audio production capacity; maintaining consistent brand voice across campaigns

Primary needs

Natural-sounding Taiwanese Mandarin TTS for commercial use; scalable to large volumes; ability to clone and store brand voices; easy integration into workflows

Scout Review

Written by AI from measured evidence, scored out of 100.

Soundwaver is an ambitious AI TTS platform aimed at Traditional Chinese content creators, offering voice cloning, Director Mode, and a community voice library. The design is charming but flawed (red scribbles, dense testimonials), usability is strong with clear onboarding and docs, but speed is a major issue (LCP 20.8s on mobile). Security is decent with HTTPS and consent policies, but lacks CSP and certifications. Accessibility is good but not perfect. Growth potential is high given the underserved niche, but the product is very early-stage. Overall, a promising tool with execution gaps to address.

Measure by measure

Usability

Clear three-step onboarding, structured docs with interactive tag playground, and free tier without credit card make first-time use approachable.

Design

Consistent hand-drawn style with yellow accents and clear voice-picker mockup, but red scribbles over hero headlines and dense testimonial card undermine polish.

Speed

Lighthouse mobile performance 55/100 with LCP 20.8s and FCP 18.8s — very slow initial load on mobile; desktop better at 63/100.

Security

HTTPS, HSTS, frame protection, and DMARC in place; consent policy for voice cloning is solid. Missing CSP and Permissions-Policy, no published security certifications.

Accessibility

Lighthouse accessibility 90/100 with minor issues: aria-* mismatches, contrast problems, missing main landmark. Bilingual interface helps non-Chinese speakers.

Growth

Targets underserved Traditional Chinese TTS niche with clear differentiators (Director Mode, voice cloning, community library) — real headroom if execution holds.

Improvement areas

Mobile performance: LCP 20.8s is unacceptable; optimize assets and use CDN.
Design polish: remove red scribbles from hero, improve testimonial layout spacing.
Security transparency: publish security page, add CSP and Permissions-Policy headers.

Design & Usability

Soundwaver's design is a mixed bag: consistent hand-drawn style and yellow accents give it a friendly, approachable feel, and the voice-picker mockup is clear. However, red scribbles over the hero headlines (both Chinese and English) look like an editing mistake, and the dense testimonial card hurts readability. Usability is stronger: the three-step onboarding is intuitive, the docs are structured like a game manual with interactive tag playground, and the free tier requires no credit card. The bilingual interface is a plus, though the docs are only in Traditional Chinese.

Speed & Security

Performance is a weak point: Lighthouse mobile scores 55/100 with LCP at 20.8s, which is painfully slow for a landing page. Desktop is better at 63/100 but still not great. Security is more solid: HTTPS, HSTS, frame protection, and DMARC are in place, and the FAQ clearly states voice cloning requires consent. However, missing CSP and Permissions-Policy headers, and no published security certifications (SOC 2, ISO) leave room for improvement. The consent policy is a good ethical touch.

Accessibility & Growth

Accessibility is decent: Lighthouse scores 90/100, but there are real issues like aria-* mismatches, insufficient color contrast, and a missing main landmark. The bilingual interface helps, but no WCAG statement is published. Growth potential is promising: Soundwaver targets Traditional Chinese content creators, a niche underserved by ElevenLabs and Google TTS. Its differentiators—Director Mode, voice cloning from 30-second samples, and a community voice library—are concrete. The domain is only 0 years old, so traction is unproven, but the idea has clear headroom if execution improves.

Conclusion

Soundwaver has the right idea for a niche market, but it needs to polish its design, fix the hero scribbles, and dramatically improve mobile load times. The docs and free tier are great for onboarding, and the consent policy is a good ethical foundation. If the team can address speed and accessibility issues, and build community traction, it could become a go-to for Traditional Chinese TTS. For now, it's a promising early-stage product worth watching.

Suggestions for improvement

Remove the red scribbles from hero headlines; they look like a design error.
Improve mobile performance: optimize images, lazy-load, and consider a CDN to cut LCP below 3s.
Add a proper security page with details on data handling and any certifications (SOC 2, ISO).
Publish an accessibility statement and fix contrast and landmark issues flagged by Lighthouse.

How we score

Alternatives comparison

Named competitors, point by point. Nobody paid to appear here or to be left out.

	Soundwaver	ElevenLabs	Microsoft Edge	Google Cloud Text-to-Speech
Traditional Chinese voice quality	Specialized in Traditional Chinese with local accents (Taiwanese, etc.)	Multilingual but less natural Mandarin/Taiwanese accents	Built-in TTS with multiple voices but limited customization	Robust API but not optimized for Traditional Chinese
Voice cloning	Clone from 30-second sample; included in free plan (1 clone)	Voice cloning available but requires paid plan	No voice cloning	Custom voices via API but requires technical setup
Emotional control	Director Mode with plain-language instructions and emotion tags	Limited emotional control; mainly via voice settings	No emotional control	Limited; requires SSML for some effects
Pricing (entry level)	Free plan with 1,000 credits; paid from $5/mo	Free tier with limited characters; paid from $5/mo	Free (built-in)	Pay-as-you-go, no free tier for TTS
Target audience	Traditional Chinese content creators (Taiwan/HK)	Global creators, multilingual	General users needing basic TTS	Developers and enterprises

ElevenLabs

Powerful AI voice generator with multilingual support, but often has less natural Mandarin/Taiwanese accents and lacks specialized Traditional Chinese features.

Microsoft Edge

Built-in TTS offers multiple voices, but limited customization and lacks voice cloning or advanced emotional controls.

Google Cloud Text-to-Speech

Robust API with diverse voices, but not optimized for Traditional Chinese (focused on simplified) and requires technical integration.

Comparing options? See Soundwaver alternatives, scored side by side

View all AI Generations