Murf AI: What It Is, Pricing and How to Use It (2026 Guide)
murf.aiMurf AI turns scripts into studio-style voiceovers for videos, courses and ads. Here is what it does, what the plans cost, and how to make your first voiceover.
PromptWises is reader-supported. If you sign up through links on this page we may earn a commission, at no extra cost to you. This does not influence what we write.
What is Murf AI?
Murf AI is a text-to-speech studio for making voiceovers. You paste or type a script, choose from a library of AI voices, adjust how each sentence is spoken, and lay the result on a timeline alongside video, images and background music. The output is an audio file or a finished video you can drop into a course, an ad or a product demo.
Murf is made by Murf Inc., based in Salt Lake City, and runs in the browser at murf.ai. Alongside the Studio it sells Murf Dub for translating and re-voicing video, a developer API built around its Falcon and Gen2 speech models, a voice changer that converts your own recording into an AI voice, and add-ons for Canva, Google Slides and PowerPoint.
The problem it solves is the gap between “I have a script” and “I have a narrated video”. Instead of recording, editing out mistakes and syncing audio to visuals, you edit text. If a line changes, you change the text and re-render one sentence.
Who it is for
- E-learning and L&D teams producing training modules and course videos that need frequent updates.
- Marketers and agencies making explainer videos, product demos and social ads on a schedule.
- Sales and customer teams recording IVR prompts, phone menus and on-hold messages, helped by Murf’s telephony export formats.
- YouTube creators who prefer not to use their own voice or who publish in several languages.
- Teachers and small businesses who want a presentable voiceover over slides without buying equipment.
Murf is not the best fit if you need the most expressive or emotional voices available, want self-serve voice cloning, or are building a real-time voice agent. For those, ElevenLabs is stronger, and we compare the two below.
Key features
- Murf Studio. A browser editor where your script is split into blocks (paragraphs) and sub-blocks (sentences). Each block appears on a timeline at the bottom, next to media and music tracks, and you preview single sentences, blocks or the whole project with play buttons.
- Voice library. More than 200 voices across 30 or more languages and accents, filtered by language, gender, age group, voice style and use case. Voices marked Gen2 use Murf’s newer model with extra controls.
- Per-block voice controls. For any block you can change tone or style, speed, pitch, pauses, and pronunciation. Gen2 voices add Emphasis (stress chosen words), Variability (generate several takes of a sentence and pick one) and Say It My Way (record or upload yourself reading a line and the AI voice copies your delivery).
- Pronunciation Library. Save custom pronunciations for brand names and jargon so they are correct in every project.
- Media and stock library. Upload video, images and music, or pull royalty-free stock media from inside the editor. Sync images and clips to voice blocks on the timeline.
- Script import. Type, paste (with options to split by paragraph, by sentence or not at all) or import DOCX, TXT and SRT files.
- Audio to Text. Transcribe audio or video inside Murf (Business plan and above), useful for re-voicing an existing recording.
- Export. Audio as MP3, WAV, FLAC or telephony a-law and u-law; video as MP4 or MOV with embedded, burned-in or separate subtitles; the script as a text, SRT or VTT file.
- Murf Dub, API and integrations. Separately priced dubbing for video, a text-to-speech and voice-changer API, and add-ons for Canva, Google Slides, PowerPoint and Windows apps.
Pricing
Murf prices the Studio, the API and Murf Dub separately. The table below covers Studio plans at the time of writing, shown at the yearly-billing rate that Murf displays by default; the help centre says yearly billing is 33% cheaper than monthly, so expect monthly billing to cost roughly a third more. Check the pricing page for the exact monthly figure.
| Plan | Price | Voice Generation Time | Projects | Seats | What you get |
|---|---|---|---|---|---|
| Free | $0 | 10 minutes | Limited | 1 | Full Studio to preview; no downloads; no commercial rights |
| Creator | From $19/month billed yearly | 24 hours a year | 100 | 1 editor | All export formats, commercial rights, Canva add-on, email and chat support |
| Business | From $66/month billed yearly | 96 hours a year | 500 | 1 editor | Everything in Creator plus Gen2 controls (Emphasis, Variability, Say It My Way), 48 hours of transcription, Business License, PowerPoint and Google Slides add-ons, priority support |
| Enterprise | Custom | Unlimited | Custom | 5 or more editors, unlimited viewers | Shared projects, access control, SSO, AI translation, deletion recovery, custom voice cloning, dedicated support |
Voice Generation Time is deducted each time a sub-block is rendered, so re-rendering after a text edit costs the length of that sub-block again. Changing voice settings without changing text is free. Unused time carries forward to the next billing cycle once, and is lost if you cancel. Extra time means upgrading or asking support to re-purchase your plan.
The API is sold on its own: a free trial with $10 of credit, then pay-as-you-go at $0.01 per 1,000 characters for the Falcon model, $0.03 per 1,000 characters for Gen2, $0.10 per minute for the voice changer and $0.02 per 1,000 characters for translation, with purchased credit that never expires. Murf Dub has its own pay-as-you-go and enterprise pricing; check the Dub tab of the pricing page.
How to use Murf AI: step-by-step
1. Sign up and open the Studio
- Go to murf.ai and click Sign Up. Email or Google works and the free plan needs no card.
- From the dashboard, create a new project. Give it a name you will recognise later, because projects count against your plan limit.
- The Studio opens with the text editor in the centre, the voice and media panels on the left, and the timeline across the bottom.
2. Add your script the right way
Click into the first block and either type, paste, or use Import Script File for a DOCX, TXT or SRT file.
When you paste, Murf shows a Paste Preference dialog. Choose Split Script by Paragraphs for longer scripts, Split Script by Sentences when you want fine control over each line, or Don’t Split for a short ad read that should flow as one take. Each paragraph becomes a block and each sentence a sub-block, which is the unit Murf renders and bills.
To tidy the structure, hover between blocks and click Add a block, place the cursor between two blocks and click Merge, or split a sub-block by hovering on its left edge. Name important blocks by clicking the Block label in the top-left corner of the block.
3. Choose a voice and set the delivery
- Open the Voices and Language Library and filter by Language, Gender, Age Group, Voice Style and Use Case. Match the use case to the content: an e-learning voice sounds wrong in a 15-second ad and the reverse is also true.
- Preview a voice on your own text before applying it. Apply it to one block, then use the block menu’s Apply Settings to Project to copy voice, style, speed and pitch to every block.
- Select a block to open its settings. Adjust Speed and Pitch in small steps; add Pause markers where a breath would fall; open the Pronunciation panel to respell any word that sounds wrong and save it to your library.
- On Business and above with a Gen2 voice, select words and apply Emphasis, use Variability to generate alternative takes of a tricky sentence, or record the line yourself with Say It My Way so the AI mirrors your timing.
4. Render and listen
Click the empty play button beside a sub-block to render it. The button turns black when the audio is ready. Render block by block as you edit rather than rendering the whole project each time, since every render of changed text uses Voice Generation Time. The project-level play button on the timeline plays everything in sequence.
5. Add visuals and music
- Click Upload Media to add your own video or images, or browse the Stock Media Library for royalty-free clips, images and background tracks.
- Drag media onto the timeline. Each voice block has a matching slot, so you can sync an image or clip to the sentence it illustrates. Use the timeline’s zoom and volume controls to tune the mix, and add a fade on the music track from the track options.
- For a background track that starts before the voice, add it to the beginning of the timeline and offset the first voice block.
- In Project Settings, set the aspect ratio and resolution before you export a video, and enable captions if you want them generated from the script.
6. Export
Open the Export menu (paid plans only). Choose Voice for audio in MP3, WAV, FLAC or a-law and u-law, with Low (8 kHz), Medium (24 kHz) or High (48 kHz) quality and mono or stereo. Choose Video for MP4 or MOV, ticking Include subtitles in the video for toggleable captions or Burn subtitles to the video for permanent ones. Voice + Music exports the mixed audio and Script gives you the text, SRT or VTT. The Download Manager keeps previous exports. If you used a stock soundtrack, Murf generates a YouTube Code to prevent Content ID claims.
7. Share for review
Every plan can create a Preview Link so a client or colleague can listen in the browser before you export. Enterprise workspaces can share projects for editing with roles and access control.
Tips to get better results
- Write the script for speech first. Short sentences, contractions and no parentheses. Murf reads what you give it, so an email-style script sounds like an email.
- Split by sentences for anything with pacing. It costs nothing extra and lets you tune one line without re-rendering the paragraph.
- Fix pronunciation once in the library. Product names and acronyms are the most common complaints, and the library fixes them across every project.
- Use pauses instead of punctuation tricks. Murf’s pause markers are more predictable than stacking commas or full stops.
- Keep speed and pitch near default. Small adjustments sound natural; large ones expose the synthesis.
- Audition voices with your own hardest line. A voice that handles your technical sentence well will handle the rest.
- Render as you go. Rendering one sub-block at a time catches problems early and keeps your Voice Generation Time for the lines that need it.
- Export audio and video separately for serious projects. Murf’s video tools suit simple slideshows; for anything heavier, export the audio and finish in a video editor.
Limitations and common problems
- Free plan cannot download. You can build and preview everything, but exporting needs Creator or above. Plan your trial as an evaluation, not a production run.
- Voice Generation Time is finite and carries over once. Repeated re-rendering of long blocks after small edits drains it. Edit, then render.
- Gen2 controls are Business-only. Emphasis, Variability and Say It My Way are not on Creator, which limits how much you can shape delivery on the cheaper plan.
- Voice cloning is not self-serve. It requires a studio recording session of one to two hours per style and an Enterprise agreement with a turnaround of one to four weeks.
- Language coverage is inconsistent in Murf’s own materials. Different pages say 30+, 33, 35+ and 40+ languages. Check the voice library for the language you need before buying.
- Single editor seat below Enterprise. Creator and Business include one editor, so teams either share a login against the terms or pay for Enterprise.
- Expressiveness. Murf voices are clear and professional but flatter than the best expressive models elsewhere. For storytelling or character work, test ElevenLabs as well.
Alternatives
- ElevenLabs - more natural and expressive voices, self-serve cloning and a wider product range, but no timeline for media. See our ElevenLabs guide.
- WellSaid Labs - built for enterprise e-learning and training content, with strong pronunciation control and team features.
- Play.ht - large voice catalogue and a generous API, popular with podcasters and app builders.
- Synthesia - if you want an AI presenter on screen rather than only a voice, this is the comparable tool for training video.
- Descript - a full audio and video editor with built-in voice cloning, better when you edit real recordings as well as synthetic ones.
See more options in our AI content and media tools directory, and read our guide on how to write better prompts if you draft scripts with ChatGPT or Claude before bringing them into Murf.
Verdict
Murf AI is the right tool for teams that turn scripts into finished training, marketing and support audio every week and want one editor that handles voice, visuals, music and captions. Its block-based controls make the output predictable and easy to revise, and the telephony exports are a real advantage for IVR work. It is not the tool for the most lifelike voice, for cloning your own voice without a contract, or for occasional users who will never justify the Creator plan. Use the free plan to test a real script, then buy Creator for solo work or Business if you need the Gen2 controls and transcription.
Frequently asked questions
Is Murf AI free?
There is a free plan with 10 minutes of voice generation time and 10 minutes of transcription, no credit card needed. It lets you build and preview projects but not download them. Downloads and commercial rights need a paid plan, which starts with Creator at $19 a month billed yearly at the time of writing.
What is Voice Generation Time in Murf?
Voice Generation Time (VGT) is Murf's usage credit. Every time you render a sub-block of text to speech, the length of the audio is deducted. Changing the voice, speed, pitch or pause for the same text costs nothing. A 1,000-word English script uses about six minutes.
Can I use Murf voiceovers commercially?
Yes on paid plans. Creator includes commercial rights for individuals; Business adds a Business License for registered companies. The free plan has no commercial rights.
Does Murf AI support voice cloning?
Only as an Enterprise service. Murf prepares a script, you record one to two hours of studio-quality audio per voice style, and Murf builds the voice in one to four weeks. There is no self-serve cloning.
What formats can Murf export?
Audio exports are MP3, WAV, FLAC and a-law or u-law WAV for telephony, at 8, 24 or 48 kHz in mono or stereo. Video exports are MP4 or MOV with optional embedded or burned-in subtitles, plus SRT or VTT caption files.
How many languages does Murf have?
The pricing table lists 200 or more voices across 30 or more languages and accents, including several English, Spanish, French, Portuguese and Chinese variants. Not every voice supports the newer Gen2 features.