Voice Changer for Hollow Knight: Silksong — Complete Setup Guide
Hollow Knight: Silksong has been one of the most anticipated indie games in years. Team Cherry’s follow-up to their 2017 underground classic puts the warrior-princess Hornet at the center — a graceful, sharp-tongued fighter navigating a new bug kingdom full of strange creatures, lethal enemies, and layered lore. Whether Silksong is freshly in your hands or you are still waiting for that release confirmation, the community around it has already grown into a creative force: fan art, lore speculation, Let’s Plays, and roleplay streams are part of the Hollow Knight universe long before the credits roll.
A silksong voice changer sits right at the intersection of that creativity. This guide covers how to use voice-changing software to voice Hornet, build a cast of bug-kingdom NPC voices, narrate Let’s Play content in character, and stream it all live through OBS and Discord with zero audio mess.
TL;DR
- Hollow Knight: Silksong’s Hornet voice profile is a crisp soprano — achievable with modest pitch shift and formant tuning
- A virtual microphone from your voice changer routes directly into OBS, Discord, and any streaming platform
- Hotkey preset switching lets you move between Hornet, NPCs, and your narrator voice mid-stream
- AI voice cloning lets you build a consistent, re-usable Hornet-style persona for long-form content
- No kernel driver, no anti-cheat conflict — single-player games have no detection systems
- VoxBooster runs on Windows 10/11, low-latency audio capture-based, under 300ms latency
Why the Hollow Knight Community Loves Voice Work
The original Hollow Knight used a musical-chirp language for its characters and relied almost entirely on text for narrative. Fans filled the silence. Decades of voice-acting fan projects, animated lore videos, and reading-in-character Let’s Plays have given the community a deep habit of creative voice interpretation.
Silksong shifts some of that energy by giving Hornet actual voiced moments and a wider cast, but the indie game still leaves enormous space for community interpretation. That means there is genuine audience appetite for:
- Let’s Plays where the streamer voices Hornet and reacts in character
- Lore-explanation videos with NPC voices added for atmosphere
- Discord community roleplay sessions set in the Kingdom of Pharloom
- Fan audio dubs of the game’s text
- ASMR-style “bug kingdom NPC encounter” content
A voice changer makes all of this more achievable. You do not need to be a trained voice actor — you need software that meets you halfway and handles the heavy lifting.
Understanding Hornet’s Voice Profile
Before you dial in settings, it helps to understand what Hornet’s voice actually is. She is:
- A soprano range speaker — lighter, brighter, and higher than an average speaking voice
- Controlled and crisp — each word is precise, slightly formal, not breathy or casual
- Carrying authority — graceful but with underlying steel; she is a warrior, not a courtier
- Minimal affectation — the voice does not tip into parody or exaggeration
This is an unusual target. Most voice changer use-cases aim for deep monster voices or robotic filters. Hornet requires subtlety: a modest lift in pitch, a reduction in chest resonance, and just enough presence in the upper mids to sound clear and authoritative.
Recommended starting settings
| Parameter | Starting Value | Notes |
|---|---|---|
| Pitch shift | +2 to +4 semitones | Lifts without going into falsetto |
| Formant shift | +1 to +2 | Reduces chest weight, keeps vowel clarity |
| High-pass filter | 120–150 Hz | Removes low mud while keeping body |
| Reverb | Very light (5–10%) | Slight ethereal quality, not hall reverb |
| Noise gate threshold | -40 dB | Keeps game audio from bleeding through mic |
| Output gain | Match natural level | Prevent clipping at the virtual mic output |
These settings work as a foundation. The right exact values depend on your natural voice — someone with a naturally higher voice may need +1 semitone at most; a lower natural voice may need +5 or +6.
Building a Bug-Kingdom NPC Voice Roster
Silksong introduces a cast of insectoid characters, traders, guards, and strange entities scattered through Pharloom. For Let’s Play streamers and fan-content creators, giving each a distinct voice transforms the experience for your audience.
Hornet (protagonist)
Your main preset. Soprano, crisp, authoritative. Settings above.
The Pale Watcher / Elder NPCs
Deep, slow, weighted. These characters carry the kingdom’s history. Use:
- Pitch shift: -3 to -5 semitones
- Formant: -1 to -2
- Reverb: 20–30% with medium tail
- Slight low-end EQ boost around 200 Hz
Merchant and Trader Characters
Warmer, more casual — these characters deal in commerce and banter. Use a slightly lowered pitch (-1 semitone), increased mid-range presence, minimal effects. The goal is “friendly shopkeeper” not “villain.”
Guard and Soldier NPCs
Clipped, formal, slightly mechanical. A light bitcrush effect (8-bit degradation applied very subtly) suggests insectoid precision without going full robot. Pitch around -2 semitones from natural.
Hollow / Void-Touched Entities
These are the unsettling ones — hollow voices with no emotion. Use:
- Pitch: -6 to -8 semitones
- Formant: -3
- Chorus effect: slow, wide
- Reverb: large hall, 40–50%
- Slight distortion or overdrive for texture
Unknown / Mysterious NPCs
A slight vocoder or telephone band EQ (boost 1–4 kHz, cut below 300 Hz) creates a sense of something filtering through, not quite real. Effective for Silksong’s stranger encounters.
Hotkey Layout for Live Streams and Let’s Plays
The practical goal is to move between your narrator voice and character voices without touching the mouse. Here is a functional layout for a Silksong Let’s Play streamer:
| Key | Voice |
|---|---|
| F1 | Your natural narrator voice |
| F2 | Hornet (main character) |
| F3 | Elder / Pale NPC |
| F4 | Merchant / Trader |
| F5 | Guard / Soldier |
| F6 | Hollow / Void entity |
| F7 | Mysterious / Unknown NPC |
| F8 | Soundboard trigger (bug ambience, UI sounds) |
With this layout you can react to a new character in the game, tap a function key in under a second, and voice the response without any visible interruption in the stream. Chat reads it as fluid. It takes a few practice sessions to build the muscle memory, but most streamers have it locked in within a week.
OBS Setup: Routing the Voice Changer to Your Stream
OBS does not process your voice — it just captures it. The correct signal chain is:
Physical mic → Voice changer software → Virtual microphone → OBS
Step-by-step for Windows
- Install your voice changer (VoxBooster, for example, creates a virtual mic automatically on install)
- In OBS, go to Settings → Audio → Mic/Auxiliary Audio and select the virtual microphone as your input device
- Add a Mic/Aux audio source in your scene — it will show your processed voice in the audio mixer
- Set the voice changer’s output to match OBS’s expected sample rate (48 kHz is standard for streaming)
- Test with the OBS audio meter — you should see your processed voice at -12 to -6 dBFS peak
VoxBooster uses low-latency audio capture, which is Windows’ native low-latency audio API. This means the virtual mic appears reliably in every application without driver installation.
Monitoring your own voice
Enable monitoring in your voice changer to hear your transformed voice through headphones during the stream. This lets you adjust your vocal performance to complement the effect. Turn monitoring off before going live if you are not using headphones — you will create a feedback loop through the mic.
OBS audio sync
If you notice your voice arriving slightly behind your video, add a small delay offset to your video capture source (not the audio). 100–200ms of video delay typically syncs a low-latency audio capture voice-changer chain without further adjustment.
Discord Setup for Community RP Sessions
The Hollow Knight community has active Discord servers where members run lore roleplay sessions, listen to fan audio, and react to Silksong news together. Running a voice changer in Discord lets you voice characters live in voice channels.
Setup
- In Discord, go to User Settings → Voice & Video
- Set Input Device to your voice changer’s virtual microphone
- Disable Discord’s noise suppression — your voice changer handles this, and Discord’s processing can degrade the AI-processed signal
- Set Input Mode to Voice Activity with a threshold low enough to catch your voice but above background noise
RP session tips
- Create a text channel alongside the voice channel where other participants can read the “script” or RP prompts
- Announce which character you are switching to before switching — helps the audience follow in text even if the voice is convincing
- Record sessions with OBS or Craig bot for later fan edits
AI Voice Cloning for Consistent Long-Form Content
If you plan to create ongoing Silksong content — a serialised Let’s Play, a fan audio drama, a weekly lore analysis — consistency of character voice matters. Tap F2 in episode 1 and episode 40 and Hornet needs to sound like the same person.
AI voice cloning addresses this. You train a model on your own recorded voice performing as Hornet, and the software reproduces that timbre reliably across sessions, microphones, and room conditions.
How to train a voice model for Hornet
- Record 5–10 minutes of your own voice performing in the Hornet style — crisp, precise, slightly elevated pitch, controlled
- Keep recordings in a quiet room, consistent mic position, consistent level
- Load the recordings into the voice cloning training interface
- After training (typically 10–30 minutes depending on hardware), test the model live
- Save the model as a preset and assign it to your F2 hotkey
This is different from impersonating the actual game character — you are training a voice that you perform, shaped by you, not a copy of any game audio. This approach is legally clean and produces a distinctive, original interpretation of the character.
VoxBooster’s AI cloning processes in real time on your GPU with sub-300ms latency, which is fast enough for live streaming without noticeable delay in lip-sync.
Soundboard Integration: The Bug Kingdom Ambience Layer
Hollow Knight’s audio design is renowned. Team Cherry layered orchestral strings, dripping cave sounds, and ambient insect textures into a masterwork. Adding matching soundboard layers to your stream contextualizes the voice work.
Recommended soundboard clips for Silksong content:
- Cave drip / stone echo: passive ambient underneath all character voices
- Silk movement sounds: whoosh and light impact for Hornet action lines
- Bell tones: for transitions between scenes or lore explanation segments
- Crowd murmur (insectoid): for town and settlement encounters
- Void resonance: low drone for hollow entity voices or dark revelation moments
Bind these to your F8+ keys in a separate soundboard preset. Trigger them just before speaking a character line — the combination of sound and voice change lands as cohesive and produced even in a live stream context.
Let’s Play Narration Style: Staying in Character
Voicing Hornet during a Let’s Play requires a split focus: you are reacting to gameplay in real time while maintaining a voice character. A few approaches that work:
Hybrid narration: Your natural voice handles meta commentary (“I need to save here,” “This boss is rough”). Tap F2 and shift to Hornet-voice for in-character reactions to lore moments, story beats, and important NPC encounters.
Full character mode: Everything in character. You never break — even deaths and retries are voiced as Hornet’s determination. Harder to sustain but generates highly shareable clips.
Voice-over edit mode: Record gameplay live without voice, then record voice-over in character and sync in editing. Allows perfect vocal performance with no pressure during gameplay. Not technically live, but produces high-quality YouTube content.
Most successful Silksong Let’s Plays mix the first and third approaches: live hybrid narration for Twitch, edited voice-over version for YouTube.
Technical Comparison: Voice Changer Options for Silksong Content
| Feature | VoxBooster | Voicemod | MorphVOX | Clownfish |
|---|---|---|---|---|
| low-latency audio capture virtual mic | Yes | Yes | Yes | Yes |
| AI voice cloning | Yes | Yes (limited) | No | No |
| Hotkey presets | Unlimited | Limited (free tier) | Yes | Basic |
| Soundboard built-in | Yes | Yes | No | No |
| Kernel driver required | No | No | Yes | Yes |
| Windows 11 compatible | Yes | Yes | Partial | Partial |
| Latency (AI mode) | < 300ms | ~300ms | N/A | N/A |
| Starting price | $6.99/mo | $7.99/mo | $39.99 one-time | Free |
For Silksong content creators, the key differentiators are AI cloning (for Hornet character consistency), a built-in soundboard (for ambient layers), and no kernel driver (for stability alongside game capture in OBS).
Community Content Ideas: What Works for Silksong Audiences
The Silksong community has been waiting for years. Audiences are passionate, lore-focused, and highly engaged. Content ideas that resonate:
- “Hornet explains” format: short videos where you voice Hornet explaining a lore concept, cut with game footage
- Fan dialogue: voiced readings of community-written dialogue between Silksong and Hollow Knight characters
- First-reaction Let’s Play: play blind, voice Hornet’s reactions live — authenticity reads well
- Silksong lore speculation: voice multiple bug-kingdom characters in a “council debate” format
- Bug kingdom travel guide: comedic format, you as narrator voice several NPC locals
Each of these formats benefits directly from a reliable voice changer setup. The voice does not need to be perfect — it needs to be consistent and committed.
Frequently Asked Questions
Can I use a voice changer for Hollow Knight Silksong on console? Voice changers require software running on a PC. If you are capturing console gameplay through an HDMI capture card into OBS on Windows, you can add the voice changer to your PC microphone chain and it works normally. The game runs on console; your voice runs through the PC audio chain independently.
Does the Silksong release date affect how I should set up my content? Set up your rig and practice before release. Many streamers who had their voice preset layout and soundboard ready before a major launch captured significantly more audience in the first 48 hours, when search traffic is highest and first-impressions content is what viewers are actively seeking.
Hollow Knight: Silksong is developed by Team Cherry. Learn more at the official Team Cherry website and on the Hollow Knight Wikipedia page. VoxBooster is an independent third-party tool with no affiliation with Team Cherry.