Multilingual Voiceover Guide: From Chinese Video to English Version
Multilingual voiceover is retelling, not translation. Three approaches, a six-step workflow, decision tables, and four common traps.
Read →Soundwaver · BLOG
Soundwaver Blog Field Notes
Practical notes on AI text to speech: choosing tools, sounding natural, and turning voice into an advantage.
43posts
6topics
Latest
Podcast AI Voiceover Workflow: Build a Show with AI
No good voice, no equipment? A seven-step workflow to plan, script, voice, edit, and publish a podcast with AI.
Read →All posts
42
AI Voiceover News: ElevenLabs v4, Microsoft MAI Models, Real-Time Voice Agents
ElevenLabs v4, Microsoft MAI audio models, and a doubled valuation: what happened in voice AI this week, and what creators should do.
Read →
Voice Branding Guide: Turn AI Voiceover into a Brand Asset
A practical voice branding guide: why brands need a signature voice, four dimensions, a five-step setup, and how AI voiceover makes it affordable.
Read →
Taiwanese Mandarin AI Voice Guide: Particles, Tones, Vocabulary
Taiwanese Mandarin is a speaking habit, not a character set. Four checkable tells, three test lines, and five common fake-accent traps.
Read →
AI Voiceover for Online Courses: What Stays Human
A real case where an all-AI course halved completion. Seven-step mixed workflow, AI ratios by course type, and what keeps students listening.
Read →
AI Audiobook Production Workflow: From Text to Deliverable
The hard part of AI audiobooks is lowering redo cost. Covers listenable text prep, voice strategy, chapter archiving, long-paragraph handling, and delivery checks.
Read →
AI Voiceover Cost Control: How to Calculate Before You Buy
Three formulas to calculate AI voiceover cost: monthly minutes, cost per minute, and payback. Includes hidden fees and a budget template.
Read →
YouTube AI Voiceover SOP: A Repeatable Workflow (Templates + Checklist)
Turn AI voiceover into a repeatable process: three script templates, five writing rules, segment-based generation, and an 8-item pre-publish checklist.
Read →
AI Speech Generation Explained: Tools, Cost, and Workflow 2026
Plain-language intro to AI speech generation: how it works, three modes, pricing math, license checks, and a 30-minute plan.
Read →
AI Voiceover Complete Guide 2026: 7 Steps to Usable Narration
Write, pick a voice, test, fix, license, and publish. Includes test scripts, fix table, and a 30-minute plan.
Read →
Voice Cloning Guide: Make Your Own AI Voice (Steps, Cost, Consent)
Recording prep, cloning steps, pricing math, and consent rules in one guide. Includes test scripts, decision table, and FAQ.
Read →
Free Chinese TTS Compared 2026: 6 Text-to-Speech Tools Tested
Same real script, six free Chinese TTS tools scored on naturalness, Taiwanese Mandarin, free quota, and commercial rights. Decision table + FAQ.
Read →
Multi-Character Dialogue TTS: Cast Every Line
Tag who speaks, give each role a voice, keep narrator and dialogue separate — with casting tips and FAQs.
Read →
Publishing an AI Audiobook: From Finished Audio to Sellable
Done generating? Mastering, rights, cover, and platforms — how to turn AI chapters into a product.
Read →
Short-Video Hooks: First 3 Seconds of AI Voiceover
Shorts are won or lost in three seconds. Hook formulas, a 15-second structure, and AI pacing tips.
Read →
Voice Cloning Consent Guide: What You Can and Cannot Do
Cloning takes 30 seconds. Consent is the part people skip. A plain-language guide to permissions, commercial use, and red lines.
Read →
How Much Does AI Voiceover Cost? Credits Explained Simply
Per second, per character, or per generation? How to compare AI voiceover prices without getting lost in billing units.
Read →
How to Pick Taiwanese Mandarin TTS (Not Just “Chinese”)
Most “Chinese TTS” sounds like a news anchor or a tourist. Here is how to listen for particles, rhythm, and word choice — and find a real Taiwanese Mandarin voice.
Read →
How to Use Text-to-Speech: Make a Usable AI Voiceover in 3 Minutes
No studio, no waiting on talent. From script to export in three minutes — with copy-paste steps and common mistakes.
Read →
AI Voiceover Blooper Reel: 10 TTS Fails I Still Laugh At (and How to Fix Them)
The AI read "bank" as a place to sleep and turned 3.5万 into a math lecture. Ten real TTS blooper categories, each with a five-second fix you can apply before you hit generate.
Read →
Is AI Voiceover Allowed on YouTube? What the Monetization and Originality Rules Actually Say
My client blamed the AI voice for 100-view videos. Four months of testing says YouTube punishes content nobody finishes. Checklist inside.
Read →
AI Voiceover for Online Courses: Production Notes from Three Instructor Projects
Three instructors, three course formats, one AI voice pipeline. Script adaptation, voice selection, and every pitfall I hit along the way.
Read →
How I Built a Podcast with AI Voice: The Full Production Workflow
No microphone, no studio, no audio engineer. I produced an entire season of podcast episodes using AI voice. Here is every step, start to finish.
Read →
How to Pick the Right AI Voice: What I Learned from Testing Fifty
Same script, fifty voices, two full days. These are the five filters that actually separate usable voices from demo-only ones.
Read →
How to Do Funny Voiceovers with AI: 50 Meme Videos Later, Here Is What Works
Comedy lives in timing, not in your script. I made 50 meme videos with AI narration — here are the voice choices, pacing tricks, and copy-paste script structures that actually landed.
Read →
10 Text-to-Speech Myths Debunked (By Someone Who Uses TTS Daily)
AI voices always sound fake. Platforms bury TTS content. Ten seconds clones anyone. I tested the ten biggest claims — here is what held up.
Read →
8 Rules for Writing Scripts AI Can Actually Read Out Loud
Same engine, different scripts: one sounds like a person, the other sounds like a vending machine. Eight tested rules for writing voiceover scripts, with before/after rewrites.
Read →
How to Mix Background Music Under a Voiceover: Levels, Ducking, and Timing
The office turned around: my BGM was only 4 dB under the voice. Real numbers inside: bed -15 dB, ducking -6 to -10, tempo, license traps.
Read →
Voiceover Sounding Off? A 12-Symptom Troubleshooting Checklist
Distortion, wrong breaks, mispronounced words, uneven pacing, mismatched volume. When the same workflow suddenly goes wrong, stop guessing—use this checklist.
Read →
Taiwanese Mandarin vs Mainland Accent in AI Voiceover (TTS Test)
The same sentence, and it just doesn't sound Taiwanese. Five test sentences and a word table that show you how to tell, in seconds, whether a voice was made for Taiwanese ears.
Read →
Commercial AI Voiceover Licensing: Three Mistakes I Made
Ads going live, videos being monetized, contracts due—then the licensing falls apart. Three real screw-ups that explain the difference between tool rights, voice rights, and finished-audio rights, plus a pre-flight checklist.
Read →
Turning a Whole Book into an Audiobook with TTS: My Stability Test
Short scripts hide every weakness. When I turned a 100,000-word manuscript into an audiobook with AI, the problems surfaced fast. Here's the full playbook.
Read →
A Real Client Project: 30 Days of AI Voiceover for an Online Course
Forty-eight videos, twelve hours of course material, one instructor who refused to go back into a studio. My first full commercial AI voiceover project, recorded honestly.
Read →
Mixing Chinese and English in One Script: A Multilingual TTS Test
Code-switching is how Taiwanese people actually talk—but AI voices often butcher it. Here's what breaks, how to prep your script, and a template you can copy.
Read →
The 12 Questions I Get Asked Most About AI Voiceover
Does it sound fake? Does it cost money? Is my text safe? Can I use it commercially? I've answered these a hundred times, so I wrote them all down.
Read →
How to Add Emotion to AI Voices (Emotion TTS): Tags, Pace & Punctuation
"Make it sound excited" doesn't work. Punctuation, pacing, and word choice do. Here's what a week of trial and error taught me about putting real feeling into AI voiceover.
Read →
AI Voiceover vs. Human Voice Actor: How I Actually Decide
Budget, turnaround, emotion, revision cycles—four dimensions that decide whether a project needs a human or an AI. Includes my real quotes, my split table, and how I explain the choice to clients.
Read →
From Article to Audio: My Three-Month Commute Listening Experiment
Three hundred saved articles I never read became twenty minutes of listening every day on the train. The full workflow, what worked, what didn't, and which articles you should never convert.
Read →
I Cloned My Own Voice with AI: 30 Voiceovers Later, Here's the Truth
Ninety seconds of sample audio, twenty minutes of waiting—now my voice can read anything I write. After thirty real voiceovers: what it's great at, what it fails at, and how to record a sample that works.
Read →
How to Make AI Voiceovers Sound Human: 8 Practical Tips
The same script can sound like a textbook or a person. The difference is punctuation, pacing, and phrasing—here is what actually changed my results.
Read →
Free TTS Tools Compared (2026): Which One Fits Your Project?
Free tiers come with strings attached: quotas, license rules, voice libraries. Here is a use-case-by-use-case breakdown, tested with my own scripts.
Read →
What Is TTS (Text-to-Speech)? How It Works — Plain Guide
You hear synthetic voices every day—navigation, assistants, video narrations. Here's how text-to-speech actually works, in plain language.
Read →