ElevenLabs
Create AI voiceovers, clone your voice and transcribe recordings with ElevenLabs. Free credits are available; commercial voiceovers require a paid plan.
Create natural-sounding voiceovers for videos, audiobooks and training. Explore the best AI voice generators, from browser studios with editable narration to voice-cloning tools and text-to-speech APIs for live applications.
Create AI voiceovers, clone your voice and transcribe recordings with ElevenLabs. Free credits are available; commercial voiceovers require a paid plan.
Create AI voiceovers with Murf’s script editor, pronunciation controls and audio exports. Its speech API uses separate character-based billing.
Speechify Studio combines AI voiceovers, video dubbing and custom voices in a browser editor, with pronunciation controls and paid audio downloads.
Create AI narration for training and business videos with WellSaid. Edit pronunciation and delivery, preview takes, then use your allowance for downloads.
Resemble AI’s free Chatterbox models create speech and custom voices on your own hardware, with multilingual options and English Turbo and Nano models.
Deepgram Aura is a text-to-speech API with seven languages, named voices, streamed audio and character-based pricing for apps, narration and voice agents.
Cartesia creates AI speech in 44 languages with Sonic 3.6. Use its Playground or API for narration, custom voices and live conversational apps.
OpenAI Text to Speech creates multilingual narration through an API, with delivery prompts, stock voices and streaming. Its Speech models retire in January 2027.
Google Cloud Text-to-Speech creates multilingual narration and spoken app responses with Chirp 3 HD, prompt-directed Gemini speech and usage-based API pricing.
Amazon Polly is an AWS text-to-speech service with multilingual voices, four speech engines, SSML controls and character-based pricing for apps and narration.
Murf, Speechify Studio and WellSaid turn scripts into narration in a browser workspace. Murf combines voice selection with pronunciation, pauses and media editing. Speechify Studio adds voiceover and dubbing tools alongside stock media; it is a separate subscription from the Speechify Reader app. WellSaid centres on reusable projects, licensed voices and team narration, with paid allowances measured in downloaded audio. These studios suit creators producing a finished audio track for a video, course or book.
ElevenLabs and Cartesia offer instant and professional voice cloning, with different recording requirements and paid-plan access. Instant clones use a short reference; professional clones train on a longer recording. Delivery controls also differ: ElevenLabs has model-specific settings and audio tags, while Cartesia’s professional clones retain their recorded pacing and loudness. Cloning access still requires permission to reproduce the speaker’s voice. A voice library is another option when a project does not need a specific person’s identity.
Multilingual text-to-speech creates new narration from a script in the target language. Dubbing starts with existing audio or video and replaces its speech with a translated version. Speechify Studio offers both voiceover and dubbing, while ElevenLabs has speech models with different language coverage and a separate dubbing workspace. A language supported by one model is not automatically available in every voice or product. Voice style, speaker handling and translated-audio editing depend on the selected service.
Cartesia and Deepgram Aura provide speech APIs with streaming routes for interactive applications. OpenAI’s speech endpoint supports generated files and streamed audio, with voice and delivery options that depend on the model. Google Cloud Text-to-Speech spans several voice families, including Chirp HD and other voices with different controls. Amazon Polly connects speech synthesis to AWS through its API, console and SDKs, with pronunciation lexicons and engine-specific speech markup. A streaming speech endpoint supplies audio; managed agent calls, phone numbers and language-model usage can be separate services and charges.
OpenAI Text to Speech review →
Resemble AI’s Chatterbox family offers downloadable speech and voice-cloning models for self-hosted use. Its multilingual and English models have different language and expression controls. The software is free under the MIT licence; installation, compute and maintenance are your responsibility. Resemble stopped selling voice generation to new customers in August 2026 while retaining support for existing accounts. Chatterbox remains available to build on, providing a separate route from a hosted voice subscription.
Free access can cover voice generation without including downloadable or commercial audio. Speechify Studio’s free plan excludes exports and commercial use; Cartesia’s free speech tier is noncommercial. Paid studios also count usage differently: Murf charges generated speech time, WellSaid measures downloaded audio duration, and Speechify Studio uses credits for voiceover, dubbing and other operations. Cartesia shares speech credits between generation and transcription, with a separate dollar balance for Managed Agents. API costs and creator subscriptions can have separate allowances, even within the same brand.
LOVO Genny is an AI voiceover and video platform. Its website and app are currently unavailable. Find voice-production alternatives here.
Text-to-speech turns written text into spoken audio. AI voice generators use speech models for this task and may add voice libraries, delivery controls, cloning or a production editor. A voiceover studio produces editable narration, while an API supplies speech inside an application.
Murf, Speechify Studio, WellSaid and ElevenLabs offer browser tools for producing narration from scripts. Their editing, voice controls and download allowances differ. Speechify Studio is separate from the Reader app used to listen to documents and web pages.
Download access depends on the product and plan. Speechify Studio lets free users create speech but requires paid access for exports. WellSaid’s trial includes a small download allowance without commercial rights. Generating or downloading audio does not by itself include permission to use it in paid work.
ElevenLabs and Cartesia offer instant and professional cloning, with different audio requirements and plan limits. Chatterbox supports reference-based cloning on your own infrastructure. Use your own voice or obtain the speaker’s permission; each product can impose additional eligibility and verification requirements.
Both are common. Murf uses generated speech duration, while WellSaid measures downloaded audio. Speechify Studio uses operation-specific credits, and speech APIs can meter characters, tokens or other model-specific units. Revisions, transcription and agent calls can draw from different allowances.
Multilingual text-to-speech reads a script in another supported language. Dubbing translates speech from existing audio or video and creates a replacement track. Speaker handling, timing and editing tools depend on the dubbing product; ordinary voice generation does not automatically include them.
Chatterbox can generate speech on your own infrastructure after model setup. Its software licence is free, while hardware and maintenance costs remain separate. Hosted voice studios and cloud speech APIs process audio through their own services; their privacy and retention options are product-specific.
Turn interviews, recordings and meetings into searchable text. Explore AI transcription services, meeting-note assistants and transcript-based editors for capturing conversations, sharing summaries and producing captions.
Turn lyrics into a song, compose an instrumental or create a soundtrack with AI music tools. Explore Suno and Udio for vocal tracks, AIVA for editable compositions, and dedicated beat and background-music generators for videos, podcasts and games.
Convert, edit and prepare files with our own browser utilities.