Record into the browser or upload the take you already have. You get session-director notes on which word carried the line, where the breath went, whether the energy sagged in the middle, and whether it sounds like a person or an announcer. Timestamped, honest, and available at the hour the audition is actually due.
You cannot hear your own read accurately while you are giving it. Your voice reaches you through bone conduction, so what you experience in the booth is not what the microphone captured. That single fact is why voice over coaching exists at all, and why practicing alone plateaus so fast. You do the rep, you feel good about it, and the client hears something else entirely.
A coach fixes that by being an outside ear. The trouble is scheduling and price. Coaching starts around a hundred dollars an hour, and the auditions that need an ear most are the ones with a four hour turnaround at nine at night. So the reads that go out are almost always the unchecked ones.
Take One is an outside ear you can use on those reads. You record directly in the browser, upload a file, or paste the copy and read it live, and the coaching comes back on the audio you actually produced. Not on the script, not on the character brief, on the take.
The analysis is scoped to what is audible, which is the whole point. It does not guess at things it cannot hear.
When a rate measurement is available, it uses the number rather than a vibe. A read described as unhurried when the words per minute say it is rushing is exactly the kind of note that destroys a coach's credibility with a working artist, so the number wins.
Every analysis opens with one of three verdicts.
It clears the bar. The notes stop there, and anything remaining is flagged as preference and not a fix.
Close, and one thing is costing you. You get that one direction properly, not a full audit.
Something fundamental is not landing, so the detailed pass is warranted and you get it.
This exists because the endless loop is what makes voice actors quit. Apply the note, send the next take, receive a fresh wall of brand new notes, never finish. When a read is good and the only remaining changes are matters of taste, it says that out loud, because you cannot tell the difference between a real fix and someone else's preference unless you are told.
Scene Partner comes with the same subscription and it is not just for on-camera work. Paste dialogue copy or animation sides, tag which role is yours, and it reads the other parts out loud while listening for your lines, picking back up the moment you deliver. It forgives paraphrasing, so it plays like running lines with a person. Cue, Fade and Speed-through drills get you off book, and every pass ends with an off-book score and the lines you dropped. No reader to book, at any hour.
This coaches audio. It cannot see you, so it never comments on posture or breathing mechanics you would want a human to catch in the room. It does not know which categories your particular voice sells in, and it has no casting relationships. A human coach is worth real money for exactly those things. What this replaces is the silence between sessions, and the 11pm audition that would otherwise go out unheard. More on that split in when you cannot book a coach weekly.
Sample Output
A conversational 30 second commercial audition, coached in full. Every note anchored to the second it happened.
Your first three seconds are warm and unhurried, and you start mid-thought rather than announcing, which is the hardest part of a conversational spot to fake. Reading this as conversational commercial, so I am judging whether it sounds like one person talking to one person. My honest read: the top is genuinely castable and the back half drifts into copy reading. What decided it was the shift at [0:18], where you stop talking to someone and start delivering.
At [0:04] you throw away "most people never check" instead of pointing at it, and that is what sells the whole conversational premise. Keep that. The pause you take at [0:09] is a real thought, not a rest, and it buys you the credibility the rest of the read spends.
"You already have everything you need to switch." You stressed need. The line wants already, because the point is that no work is required. [0:14]
"It takes about four minutes." You stressed minutes. The line wants four, since the number is the sell. [0:21]
Energy builds nicely to [0:16], then plateaus rather than continuing to move. The breath at [0:19] is taken mid-clause, which chops "everything you need" away from the reason it matters. And you push the tag at [0:28], which is audible: your volume steps up about halfway through the last line because you know it is the last line.
It tips at [0:18]. Up to there I can hear who you are talking to. After that the pitch flattens onto one plateau and the person on the other end disappears.
Before you start, decide the person you are talking to has already told you they cannot be bothered to switch banks. Then take your breath before "everything" rather than inside it, and let the last line land quieter than feels safe instead of stepping up into it.
On playback, check whether the tag is quieter than the line before it. If it is, you have it.
Self-tapes get the same treatment on the video side: eye contact, the moment before, script ownership and frame, all timestamped.
7-day trial, then $20 per month billed automatically unless you cancel. Everything included.
Start 7-Day Trial →FAQ
It listens to the audio. You record straight into the browser, upload a file, or paste a script and read it live. The coaching covers what is audible: which word carried the line and which one should have, where a breath was taken or audibly held, whether energy built or sagged through the middle, whether it sounds like a person talking or someone reading aloud in a voice, and recurring habits like upward inflection or sing-song lists. Every note is anchored to a timestamp.
Take One is one plan at $20 a month after a 7-day trial, with every tool included. A private VO coach typically runs well over a hundred dollars an hour, which is why most artists book one occasionally rather than weekly. The two are not really competing. A human coach knows your market and can open doors. Take One is what you use on the reads in between, at the hours a coach is not available.
Yes, and that is deliberate. Every analysis opens with one of three verdicts: SUBMIT THIS, ONE MORE TAKE, or NEEDS REWORK. On SUBMIT THIS the notes stop, and anything left over is labeled as preference rather than a fix. When a read is good and the only changes left are matters of taste, it says so out loud instead of dressing taste up as a correction.
The Rate Estimator is included. It works out a ballpark range for a job by classifying it by project type, usage, market, term and exclusivity, flags the details clients leave out that cost artists money, and shows the commission math at 10% and 15%. It is a sanity check rather than a rate lookup, so confirm against the current GVAA Rate Guide or your union rate card before you quote. An Agent Pitch Writer and Client Email Generator are included too.
Taping on camera as well? See online acting coaching and self tape feedback. Related reading: VO rates in 2026, submitting to voice over agents, and choosing a voice over recording app.
Pricing
7-day trial, then $20 per month billed automatically unless you cancel. No API key required. Cancel anytime.
Take One: Full Access
Cancelling takes about ten seconds and you do not need to email anyone. Manage or cancel from the homepage, enter the email you subscribed with, and Stripe ends your billing right there. You keep access through the end of the period you have already paid for. If you are stuck, write to takeonesupport@gmail.com.