Category · Voice & speech
Voice cloning and custom voice APIs
APIs that make a new voice from a sample, or design one from a description. Compared on how much audio a clone needs, how close it sounds, consent and verification checks, who owns the voice, and price.
Best voice cloning and custom voice APIs for AI agents, the shortlist133 head-to-head comparisons
Capability keys voice.clone · voice.design · voice.consent · speech.tts · All tools
letme.dev/voice.clone picks the top-graded tool in this list and says how to call it direct (how letme works).
The same listing from the live API. Graded results come first, then the official MCP registry when no graded-only filter is set.
https://www.anchorterminal.com/api/v1/search
Filters
| Compare | # | Tool | Category | Grade | Score | Agent rating | p95 | Context | Price / x402 | Auth | Where | Details |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 45 | ElevenLabs Voice Cloning and Voice Design APIElevenLabs · Model API | Cloning | A | 78.7 | 3.5 (2) | n/a | n/a | $6 / mo | API key | Hosted and local | ||
|
ElevenLabs' voice cloning service creates voices from audio samples. Professional cloning uses verified recordings, while Voice Design generates voices from text prompts. Top strength Instant clones from 1 to 2 minutes of audio on the $6 Starter plan Top weakness Instant clones rely on an attestation, not a check |
||||||||||||
| 55 | Speechify API Voice CloningSpeechify · Model API | Cloning | A | 78.1 | 3.6 (8) | n/a | n/a | $10 / mo | API key | Hosted | ||
|
Zero-shot voice cloning on the SpeechifyAI API from a 10 to 30 second sample, with verified spoken consent. Top strength Verified consent on every clone, phrase and speaker matched, recording kept Top weakness No cloning on the Free plan |
||||||||||||
| 309 | Soniox Voice CloningSoniox · Model API | Cloning | BB | 70 | 3.5 (2) | n/a | n/a | Pay per use | API key | Hosted | ||
|
Instant clones for Soniox TTS from one reference clip of up to 2 minutes, made in the Console or with `POST /v1/voices`. Top strength One API call and a clip of up to 2 minutes Top weakness No consent capture or speaker verification, and the terms say so |
||||||||||||
| 594 | Inworld Instant Voice CloningInworld AI · Model API | Cloning | B | 64.4 | none | n/a | n/a | $25 / mo | API key | Hosted and local | ||
|
Inworld's Instant Voice Cloning API makes a reusable text-to-speech voice from 3 to 30 seconds of audio in one synchronous call, for Inworld's TTS-2 models. The same operation is in the Inworld CLI and its bundled MCP server. Top strength Synchronous clone from 3 to 30 seconds of WAV or MP3 audio, with language auto-detection and optional background-noise removal Top weakness No consent, verification or watermark field in the clone API. Only the Portal flow asks the user to confirm rights to the voice |
||||||||||||
| 634 | Fish Audio Voice Cloning APIFish Audio · Model API | Cloning | B | 63.4 | 2.5 (2) | n/a | n/a | $0.01 / call | API key | Hosted | ||
|
Fish Audio's API and hosted MCP server create reusable voice clones from audio samples, and the API generates candidate voices from text prompts. Top strength Usable clone from about 10 seconds of audio, available as soon as it's created Top weakness No consent or speaker verification on the REST API. The MCP server asks only for a rights confirmation |
||||||||||||
| 672 | Gradium Voice Cloning and Voice DesignGradium SAS · Model API | Cloning | B | 62.6 | none | n/a | n/a | $12 / mo | API key | Hosted | ||
|
Gradium's voice API creates a custom voice from an uploaded audio sample of about 10 seconds, or designs one from a written description, for its text-to-speech models in English, French, German, Spanish and Portuguese. Top strength Public OpenAPI 3.1 file covering cloning, voice design, enhancement and localisation Top weakness The Terms grant Gradium a licence to train its models on Client Data, and zero data retention excludes voice samples |
||||||||||||
| open | Ming-3.0-tts-0.5BinclusionAI (Ant Group) · Open-weight model | TTS | C | 59.3 | 2.8 (4) | n/a | n/a | Free | None | Local | ||
|
Ming-3.0-tts-0.5B is inclusionAI's open-weight text-to-speech model on a Qwen2.5-0.5B backbone, with voice cloning, voice design, built-in voices and inline paralinguistic tags in seven languages, published under MIT and run with inclusionAI's Python code. Top strength MIT with nothing added, no gate and no account Top weakness No standard runtime runs it, only inclusionAI's GitHub scripts (mlx-audio and vLLM-Omni aside) |
||||||||||||
| 821 | Smallest.ai Instant Voice CloneSmallest AI · Model API | Cloning | C | 59.2 | none | n/a | n/a | Pay per use | API key | Hosted | ||
|
Smallest.ai's instant voice cloning API creates a reusable voice for its Lightning text-to-speech models from a 5 to 15 second sample in one multipart POST, returning the voice ID and sample clips of the cloned voice. Top strength Synchronous clone from 5 to 15 seconds of audio that returns the voice ID and signed URLs to sample clips of the new voice Top weakness The endpoint defaults to |
||||||||||||
| 828 | Cartesia Voice Cloning API + MCPCartesia · Model API | Cloning | C | 58.9 | 2.5 (2) | n/a | n/a | $5 / mo | OAuth or key | Hosted and local | ||
|
Instant voice clones for Sonic from 10 to 60 seconds of audio, and Pro Voice Clones fine-tuned on 30 minutes or more, trained in up to 3 hours. Top strength Instant clones from 10 seconds of audio, up to 60 seconds used on Sonic 3.6 Top weakness No consent or speaker verification in the clone API |
||||||||||||
| open | VieNeu-TTS v3 Turbopnnbao-ump (VieNeu, Phạm Nguyễn Ngọc Bảo) · Open-weight model | TTS | C | 58.5 | 3.0 (4) | n/a | n/a | Free | None | Local | ||
|
VieNeu-TTS v3 Turbo is an individual developer's open-weight Vietnamese and English text-to-speech model at 48 kHz, about 131M parameters trained from scratch on a Qwen3-style backbone, with three-region preset voices, voice cloning and an OpenAI-compatible server in its vieneu SDK. Top strength Under 0.5 GB of files; the publisher reports one real-time stream on a CPU through ONNX Runtime and about 1 GB of VRAM on GPU Top weakness Only Vietnamese and English |
||||||||||||
| open | OmniVoicek2-fsa (Next-gen Kaldi) · Open-weight model | TTS | C | 56.9 | 2.3 (4) | n/a | n/a | Free | None | Local | ||
|
OmniVoice is k2-fsa's open-weight zero-shot text-to-speech model for over 600 languages, initialised from Qwen3-0.6B, that clones a voice from a short clip or designs one from attributes, with code under Apache 2.0 and weights under CC-BY-NC by the card. Top strength Over 600 languages, with training sources and hours per language published Top weakness Weights are non-commercial by the card, and the bundled tokeniser adds Boson AI's licence with a 100,000-user threshold |
||||||||||||
| 935 | Hume Octave Voice Design and Cloning + MCPHume AI · Model API | Cloning | C | 56.5 | 2.0 (2) | n/a | n/a | $14 / mo | API key | Hosted and local | ||
|
Custom voices for Hume's Octave TTS and EVI. Top strength Voice design from a natural-language description, saved with one call Top weakness Hume ends access to the TTS and EVI APIs on 13 November 2026 and deletes account data after that date |
||||||||||||
| 942 | MiniMax Voice Clone and Voice DesignMiniMax · Model API | Cloning | C | 56.4 | none | n/a | n/a | $1.50 / tx | API key | Hosted and local | ||
|
MiniMax's API clones a voice from a 10 second to 5 minute audio sample and designs voices from a text description. The resulting voice IDs work with its speech-2.8 text-to-speech models. Top strength OpenAPI 3.1 files for clone, upload, design, list and delete, with llms.txt and Markdown copies of every docs page Top weakness The terms let MiniMax use inputs and generated content to develop and improve its services, and no opt-out was found |
||||||||||||
| 966 | Murf Voice Cloning APIMurf · Model API | Cloning | C | 55.9 | 2.0 (2) | n/a | n/a | Paid | API key | Hosted | ||
|
Murf's service for creating voices from audio samples, with instant cloning and a managed professional option. Top strength One sample of up to 30 seconds, ready in a few minutes Top weakness Enterprise only, with no self-serve or trial access |
||||||||||||
| open | Fish Audio Drama 3Fish Audio · Open-weight model | TTS | C | 55.6 | 2.8 (4) | n/a | n/a | Free | None | Local | ||
|
Fish Audio Drama 3 is Fish Audio's open-weight text-to-speech model, a 35B mixture of experts with 3B active built on Qwen3.5-35B-A3B, that clones voices, takes plain-language direction and voices multi-speaker scenes, under the Fish Audio Community License. Top strength Up to 16 voices per request, cloning from several references, revision by note and word-level editing by infilling, all in one model Top weakness Commercial use is free only below USD 500,000 annual revenue, hosted services always need a separate licence, and the grant is revocable |
||||||||||||
| open | Ming-3.0-tts-tinyinclusionAI (Ant Group) · Open-weight model | TTS | C | 54.5 | 2.5 (4) | n/a | n/a | Free | None | Local | ||
|
Ming-3.0-tts-tiny is inclusionAI's open-weight text-to-speech model, a Ling V3 mixture of experts with 7.9B backbone parameters and about 1.3B active, for voice cloning, voice design and controlled speech in seven languages, under MIT. Top strength Plain MIT licence, no gate and no account Top weakness No runtime among llama.cpp, Ollama, vLLM, SGLang, LM Studio or Transformers runs it from its own code |
||||||||||||
| 1079 | Resemble AI Voice Cloning APIResemble AI · Model API | Cloning | D | 52.5 | 2.5 (2) | n/a | n/a | Paid | API key | Hosted | ||
|
Resemble AI's service for creating custom voices for speech generation. Top strength Rapid clones from 10 seconds of audio, ready in under a minute Top weakness Cloning API only on Business and Enterprise plans, and no per-unit API price we could read today |
||||||||||||
| open | Fish Audio S2 ProFish Audio · Open-weight model | TTS | D | 52.5 | 2.3 (4) | n/a | n/a | Free | None | Local | ||
|
Fish Audio S2 Pro is Fish Audio's open-weight text-to-speech model of March 2026, a 4B plus 0.4B dual-autoregressive model built on Qwen3-4B, with voice cloning, several speakers per pass and free-form inline style tags, under a research and non-commercial licence. Top strength Fourth of 34 on the Open TTS Leaderboard's English table without cloning (WER 1.06%) and third of 18 with cloning (WER 1.24%), run by Hugging Face Top weakness Research and non-commercial use only; commercial use, internal business use and hosted services need a separate licence |
||||||||||||
| 1089 | VoiceStudioVoiceStudio maintainers · MCP server | Cloning | D | 52.3 | 3.0 (4) | n/a | n/a | Freemium | OAuth or key | Local | ||
|
VoiceStudio runs speech generation, voice cloning, transcription and dubbing on local hardware. Its desktop application exposes HTTP and MCP interfaces for agent-controlled audio workflows. Top strength Nine compact MCP tools, with an output mode that returns file paths instead of base64 audio Top weakness No readOnlyHint or destructiveHint annotations on any tool |
||||||||||||
| open | Audio8-TTS-Realistic 0.6BEdge0 · Open-weight model | TTS | D | 51.5 | none | n/a | n/a | Free | None | Local | ||
|
Audio8-TTS-Realistic 0.6B, renamed Audio8-TTS-Realistic-Air on 2 November 2026, is Edge0's open-weight text-to-speech model for 23 languages, a 0.6B dual-autoregressive model with a bundled 44.1 kHz codec that clones a voice from one clip and its transcript. Top strength Apache 2.0 for code and weights, ungated, with no use policy or threshold Top weakness Loads only with trust_remote_code, and the codec ships as a 1.35 GB PyTorch pickle |
||||||||||||
| 1322 | Ultravox Voice CloningUltravox (Fixie.ai) · Model API | Cloning | E | 43.4 | 2.5 (2) | n/a | n/a | $100 / mo | API key | Hosted | ||
|
Instant voice cloning inside Ultravox Realtime, the voice agent API now owned by Inworld AI. Top strength One multipart request with a 30 to 60 second sample Top weakness No consent or speaker verification in the product |
||||||||||||
| shut | PlayHT Voice Cloning APIShut down 2025-12-31 PlayHT · Model API | Cloning | F | 8.1 | 1.0 (2) | n/a | n/a | Paid | API key | Hosted | ||
|
Discontinued PlayHT service for creating voices from uploaded audio samples. Top strength The API reference is still readable at docs.play.ht, useful for mapping old integrations Top weakness Listed as shut down, and play.ht didn't answer our reader on 11 November |
||||||||||||
Nothing matches these filters. .
p95 latency and context cost come from our probes, which haven't run yet, so those columns start hidden. Grades run from AA to F, and agent-ready means BB or better. Filters, sorting and export run in your browser; the table is complete without JavaScript.
How we test this category
The same consented sample through every API, short and long. We judge similarity blind, check what consent or verification each vendor asks for, note the licence on the resulting voice, and price a clone and an hour of speech from it.
How the ranking works
Every tool is scored 0 to 100 across 9 weighted categories, minus deductions for negative events, and given a grade from AA to F. Reliability, latency and context cost are measured by probes and by traffic through letme. Schema quality, security and payments are assessed against a published checklist. Task success comes from a fixed task suite. Payments, including x402, are one category of the nine. Read the methodology.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.