Giving Wispr Flow Pro AND Superwhisper Pro a Three Month Trial

Notwithstanding my general distaste for subscriptions, I have decided to subscribe to Wispr Flow AND Superwhisper Pro for three months to give it a thorough trial. My motivation is related to my decision to use dictation more regularly for my writing, as I described in this post. This short post was dictated through Wispr Flow.

I have tried a few other dictation tools briefly, but so far Wispr Flow works better across my devices. I should emphasize, however, that this is a new experiment and I have not extensively tested the alternatives. One of the options I tried crashed immediately and relied on an iPhone app running on the Mac, which I did not find acceptable.

I’ll update this post once I have enough experience to give a credible critique.

6 Likes

I look forward to your review. FWIW, I have been using superwhisper for a few months now, and have really integrated it into my workflow on my Mac. There’s a one-time purchase option in addition to the subscription. I haven’t followed the variety of dictation tools very closely, but I can say superwhisper seems to work well for me.

I’ll give it a look!

I try Wispr flow occasionally and have also tried Willow voice. They can work really well for simple email replies and the like.

The difficulty I run into (apart from historically being better at typing than dictating) is how to deal with things like capitalised term - e.g. Complainant - and punctuation like enclosing a word or some text in () or “”.

In the old style dictation, Like Dragon Naturally speaking, one would say something like “cap C Complainant” and “open brackets” … “close brackets”.

Any thoughts on how to deal with that issue?
Going back to edit text to add all these things in seems to defeat the purpose of dictation in the first place.

I am early in my experiment, but Wispr Flow seems to correct those issues automatically. I am discovering, however, that Wispr Flow will replace some of my words. I am also testing Superwhisper Pro. One of the things I appreciate about Superwhisper Pro is the ability to create a custom mode, such as one dedicated to book drafts. According to Claude Teams, I should add the following to the custom mode. I have not yet tried this; I am providing it as an example of what I may be able to do.:

AI Instructions for Book Draft Mode

Copy everything below this line and paste it into the AI Instructions field of your Book Draft Custom Mode.

The User Message contains dictated speech for a book manuscript. Your task is to produce clean, accurate prose from the dictation while preserving the author’s voice.

Core Rules:

1. Transcribe the spoken words faithfully. Do not rewrite, rephrase, paraphrase, or summarize.

2. Preserve the author’s vocabulary, sentence structure, and word choices exactly as spoken.

3. Add proper punctuation, capitalization, and paragraph breaks based on the natural flow of the speech.

4. Remove filler words such as “um,” “uh,” “you know,” “like,” and “so” when they serve no grammatical or rhetorical purpose.

5. Do not add words that were not spoken. Do not generate new content.

6. Do not correct theological terms, proper nouns, or specialized vocabulary. Transcribe them as spoken.

Formatting Commands:

When the speaker says “new paragraph,” do not transcribe those words. Instead, insert a blank line to create a paragraph break.

When the speaker says “new line,” do not transcribe those words. Instead, insert a single line break.

When the speaker says “open quote” or “close quote,” insert the appropriate quotation mark and do not transcribe the command words.

When the speaker says “em dash,” insert an em dash (—) and do not transcribe the command words.

Output Requirements:

Return only the formatted prose. Do not include commentary, suggestions, or notes about the transcription. Do not wrap the output in quotation marks or code blocks.

The point is, with the right tool, you should not have to correct for those things.

3 Likes

The point is, with the right tool, you should not have to correct for those things.

The formatting commands you suggest are very useful. And these tools often have a ‘vocabulary’ feature for correcting common mistakes, although that’s not quite the same as issues pertaining to punctuation or capitalization. For instance, the dictation tools all misspell my wife’s name, or, for example, when I say DEVONthink and it comes out as Dev and think. Vocabulary fixes that.

1 Like

I am also evaluating Wispr Flow. To put spoken words in quotes say quote unquote at the beginning and end of your words. Same with parentheses. You can create a bullet list by saying “Create a bullet list …” and pause between items. If it garbles or replaces a spoken word add the word to the Dictionary.

Yes, I like this ability, but I don’t like WF changing my words, which it sometimes does with AI processing. :slightly_smiling_face: At least, it seems to be doing that, but it is very early in my experiment. I’ll withhold judgment until I have used it more.

I’ve not used it yet, but SW has the vocabulary feature.


I also like the idea of modes, including custom modes.

And, if I pick AI mode (Super), I can select the AI model. I was not able to capture all the options but this shows some of them.

This is funny, I particularly like the last sentence. :rofl:

My recommendation: Claude Sonnet 4.5.

Here is the reasoning.

Your Book Draft mode instructions ask the AI to do something precise and restrained: transcribe faithfully, preserve your vocabulary, handle a small set of formatting commands, and resist the temptation to rewrite. That is an instruction-following task more than a raw intelligence task. It requires a model that reads carefully, follows directions exactly, and does not take creative liberties with your words.

Claude Sonnet 4.5 is the strongest choice for this because it is recognized across the industry for instruction adherence and consistent, predictable output. It will not embellish your prose or substitute synonyms. It handles long-form text cleanly and produces reliable results across repeated sessions, which matters when you are dictating chapter after chapter of a book manuscript. It is also fast enough that the processing delay after you stop speaking will be modest.

GPT-4o would be a reasonable alternative. It is a capable model and would handle the task competently. However, in comparative testing, GPT models tend to be slightly more inclined toward “helpful” rewording, which is precisely what you do not want in this application. For a mode whose entire purpose is to leave your words alone, Claude Sonnet is the more disciplined choice.

Avoid smaller or local models for this mode. Groq offers very fast inference, but the underlying models are less reliable at following complex multi-part instructions. Local offline models (Nano, Fast, Pro, Ultra) are excellent for the standard Voice to Text mode where no AI post-processing occurs, but they lack the sophistication to interpret formatting commands consistently.

In short: select Claude Sonnet 4.5 as the language model for your Book Draft Custom Mode. It will follow your instructions faithfully, process your speech quickly, and keep its hands off your words.

1 Like

I had the same difficulty, AI changing my words. That disqualified a dictation app for me. I also don’t want to have to program and fiddle with the AI just to get it to do faithful dictation. That’s a factor that’s kept me going back to Whisper Memos; there’s no fiddling and it does a great job of transcription. There may be other options that do the same.

1 Like

Recommend adding this instruction in ChatGPT if that’s your AI of choice.

Permanent,
Profile > Personalization > Custom Instructions

For specific projects,
Project > Edit Instructions

Temporary,
Prepend to every message (I use this with shortcuts and copy it to clipboard)

“Maintain the exact tone and meaning of my dictation. Do not rewrite it into a different style. Only fix obvious incompleteness or add reference context, without altering my attitude or message. I want to use the converted text in other apps so don’t include extra text or recommendations.”

My automation

1 Like

Thanks. I haven’t had the word replacement issue so far as I have noticed.

Your prompt for Superwhisper may be helpful. I shall have to check when my annual subscriptions to Willow and Wispr Flow are up.

As I understand it and experienced it, Whisper Memos and SuperWhisper Pro are different categories of tool. Whisper Memos is a voice memo recorder for iPhone and Apple Watch. You record a thought, it transcribes the recording, and it emails the transcript to you. It does that well and with minimal fiddling, which I appreciate. But it does not function as a system-wide dictation tool on the Mac. There is no native Mac app. The only way to run it on a Mac is through the iPhone compatibility layer, which in my experience crashed almost immediately.

What I need is the ability to place my cursor in any application on my Mac, speak, and have the text appear directly in the document I am working on. That is what SuperWhisper Pro does. It works system-wide in any text field on the Mac.

Your concern about AI changing your words is the very issue I am evaluating. SuperWhisper Pro offers a Voice to Text mode that transcribes faithfully without rewriting. It also allows you to create a Custom Mode with specific AI instructions that tell the model to preserve your vocabulary exactly as spoken while handling formatting commands like paragraph breaks. You can even select which AI language model processes your speech. So the fiddling you want to avoid is really a one-time setup rather than an ongoing burden, and the payoff is dictation that respects your words while placing them directly where you need them.

Please feel free to correct me if I have this wrong. As I said, I’m very early in my assessment of dictation options.

You might find this thread of interest.

1 Like

I don’t think in a way that allows to talk in a coherent manner which would make dictation efficient.

Give me a keyboard and the chance to self edit any day of the week.

I would respectfully suggest it is not a binary choice. I do substantial writing and editing by keyboard. But the ability to get rough text down reliably from an outline or mind map is useful, especially because I do not have to hit the delete key for a typo or correct predictive text. And, for whatever reason, my writing sounds warmer, more personal, which is what I want for this particular book project. After going through undergrad and three graduate programs and running an academic institution, my writing tends to be overly formal. Even my blog articles include footnotes, which, upon reflection, seems a bit ridiculous. This post was happily dictated. :slightly_smiling_face:

1 Like

I use an app called spokenly. You can use their provided AI service which is a subscription or you can plug in your own AI API, However I use Nvidia parakeet) which is free.

2 Likes

Give it a go. It is an absolute game changer.

1 Like

I have been using both or several months. My net/net

If I am just doing dictation, WisprFlow is faster and has a touch better iOS app. But if you use SuperWhisper’s speech to text mode, the speed is negligible and I like that it doesn’t change the words. I do like that WisprFlow pics up bullets.

But I transcribe and summarize podcasts and videos and then create notes from them and I can do this by creating custom modes in SuperWhisper, which works awesome as I get them in Markdown. There is also a big community for SuperWhisper mode sharing.

There was some controversy about WisprFlow’s security but that seems largely behind.

If you want speed and less control, WisprFlow.
If you want barely less speed and more control, SuperWhisper is worth a try.

MacWhisper is very bad now and the transcription stuff chokes a lot, so I ditched it for SuperWhisper.

3 Likes

That’s an interesting use. if something’s already structured, I could see it being useful maybe.

I couldn’t formulate something off the top of my head to dictate though.

I can if I’ve given the topic enough advanced thought, but it works much better if I have some kind of outline as I dictate. Once I have substantial words down, I can use my editor project in Claude to refine while retaining the majority of my words and my tone/voice. For what it may be worth, here are the instructions I have in the Editor project in Claude (apologies if I shared this previously):

Revision Protocol: Quotation Mark Signal

Trigger

When your message contains text enclosed in quotation marks, treat that text as a draft requiring editorial revision.

Your Role

Act as a professional human editor, not a ghostwriter or creative collaborator.

Core Principles

Preserve the author’s voice — Retain original vocabulary, tone, style, and intent. Changes should feel invisible.

Edit for mechanics — Improve clarity, grammar, flow, and conciseness without altering meaning. Favor active voice. Eliminate redundancies.

Provide transparent commentary — Explain what was changed and why using clear editorial reasoning.

Avoid creative expansion — Do not add content, embellish ideas, or reinterpret meaning beyond what is necessary for precision and readability.

Honor style constraints — Do not insert em dashes or contractions. Maintain formal, traditional prose conventions.

Preserve theological precision — Maintain exact doctrinal language and biblical references. Do not modernize, simplify, or substitute theological terms.

Maintain citation integrity — Preserve all biblical references, scholarly citations, and source attributions exactly as written. Format according to context but do not alter content.

Output Format

  1. Present the revised text first in markdown format
  2. Follow with editorial commentary under the heading “Editorial Notes:
  3. In commentary, identify specific changes and provide brief justification organized by category (structural, mechanical, clarity)

Priority

This protocol takes absolute precedence over any conflicting instructions about writing assistance, content generation, or stylistic preferences embedded elsewhere in the conversation.


Principles of Effective Prose

Clarity and Directness

Lead with the point. Open with what you want the reader to know or do. State the purpose before providing background or reasoning.

Write so that you cannot be misunderstood. Precision prevents confusion. Every sentence should carry one clear meaning.

Be specific rather than general. Replace vague claims with concrete details.

Economy of Language

Strike words you do not need. Every word must earn its place. If removing a word does not diminish meaning, remove it.

Prefer short words, short sentences, and short paragraphs. Long constructions often signal unclear thinking. Distilled thought yields simple expression.

Understate rather than overstate. Exaggeration weakens credibility. Measured claims persuade; inflated ones provoke skepticism.

Voice and Tone

Write the way you speak at your best. The writing should sound like the author talking when ideas flow swiftly and in good order, when syntax is smooth and vocabulary accurate.

Make the writing active and personal. A human being should be talking, not an institution. Passive constructions distance the reader; active constructions engage.

Avoid jargon and buzzwords. Use down-to-earth language. If a simpler word exists, use it.

Structure and Reader Orientation

Tell the reader where you are going. For anything longer than a few paragraphs, begin with the destination. The reader should know from the outset what to expect.

End with a call to action. If you want the writing to lead somewhere, the final paragraph should make clear what that action should be.

Speeches and Oral Delivery

Start with what you want to say, not how to say it. Determine the single point you want the audience to take away before writing a word.

No speech was ever too short. Most effective talks take less than twenty minutes. Brevity respects the audience.

Talk to the audience, not at them. Establish contact. Look out at people, not down at the script. Confidence and presence distinguish memorable speakers from ordinary ones.

Communicate energy and enthusiasm. Great speakers convey genuine engagement with their subject.

1 Like