audio Tools

Scribe v3 is ElevenLabs’ most accurate speech-to-text model, for developers building on calls, meetings, interviews and media. It makes 27% fewer word errors than Scribe v2 across 90+ languages. Unlike transcription that breaks on noise, accents and crosstalk, it labels up to 32 speakers, tags laughter, music and emotion, boosts up to 1,000 of your own terms, detects 56 entity types and handles files up to 15 hours. For live use, Scribe v3 Live runs under 150ms.

mpv is a free, open-source, and cross-platform media player designed for the command line with a minimalist approach. It solves the need for a lightweight yet highly customizable player by offering advanced video scaling, hardware decoding, and extensive scripting capabilities. Built with a powerful C API, it is ideal for developers embedding media functionality and enthusiasts demanding uncompromising playback quality.

A private voice memory app for iPhone. Speak or type a thought, then ask for it later. Transcription and answers happen on the iPhone.

Designed as a set-and-forget way to control audio for every app and device on your Mac, AudioCentral delivers refined sound in a sleek, user-friendly package. Per-app volume and routing to any device, with equalizers and AI speech isolation built in. AudioCentral works directly with macOS: no kernel extensions, no reboots, no audio-stack surgery — just a clean, native experience built around the system you already use.

Hold a key, speak, and let go. Kivo inserts clean text at your cursor in any Mac app, from email to your AI agent’s prompt box. It removes fillers and resolves corrections. Audio stays on your Mac. Free, with no account. Preview: English only and unsigned.

Pauta records and transcribes meetings, interviews and voice notes, then turns them into summaries and action items. It works fully offline, keeps your data on your device, and only connects to external AI if you choose.

Nemotron 3.5 Transcribe is an open speech LLM for intelligent transcription: a single model that transcribes 40 languages verbatim or in written form, with context, speaker attribution, and word timestamps, selected with a natural language instruction. It is the best open-weights model on the English and multilingual Open ASR Leaderboards.

A hand-gesture Mac instrument and MIDI controller. Play built-in sounds and music-driven visuals, or send hand gestures straight to a compatible DAW as MIDI. Try 17 styles or the camera-free Key Deck. Free 14-day trial. Requires Apple silicon, macOS 26+.

A macOS menu bar app that pauses or lowers Apple Music while ChatGPT speaks, then returns it at the level and position you chose.

Voice Cleaner

Build a voice agent that sounds and behaves human without writing a single line of code. Design the call flow, handle objections and edge cases, connect your tools and knowledge base, and go live on a real phone number. Every call ends with clean, structured data.

AssemblyAI-TTS-1: natural speech for voice agents, streaming in 90 ms. Unlimited concurrency, no annual contract, $15 per 1M characters.

Online mastering for finished AI-generated songs. Compare eight DSP sound profiles free. Members download WAV or MP3 for $9.90 USD/month.

Yourcast is a Telegram bot for podcasts: search a show, tap to subscribe, and new episodes arrive as audio right in your chat. 2.0 adds AI summaries with timestamps you can tap to jump in the audio, questions about any episode, feeds that repair themselves, files up to 2 GB, OPML import and payments in Telegram Stars. Free: 10 episodes a week on demand. Open source (GPL v3). In English, Russian, Portuguese, Spanish, German and Hebrew.

Eleven Realtime Translation is a streaming voice layer for telecom operators, contact centers and live events. Each person speaks their own language and hears the other in theirs, across 90+ languages. Unlike caption-only tools or stitched STT, LLM and TTS pipelines, it goes from live speech to expressive translated audio over one WebSocket, keeps one consistent voice (library or your own clone), streams captions and transcripts, and lets you tune speed vs accuracy per use case.

How to Pronounce English Words using AI voice pronunciation and helpful text.

Hold fn and talk. SpeechFast types clean text at your cursor, in any app. Everything runs on your Mac. The speech model is downloaded once; after that, no audio and no text ever leaves your machine. It works offline, on a plane, behind a locked-down corporate network. Dictate in Mail, Slack, Notion, your code editor, anywhere you can type Filler words and false starts cleaned up with Apple Intelligence Your own vocabulary for names and jargon History of your recent dictations, stored locally

Speech-to-text built for conversations with more than one voice with highest accuracy than any in the industry. Lowest diarization error on several benchmarks. 45 languages, up to 25 speakers. Add-ons: subtitles, PII redaction, summaries, speaker names, translation. Better than Pyannotee, Deepgram and Assembly AI

Drama 3 is Fish Audio's latest AI voice model, built for performances you can direct. Turn scripts into expressive speech with natural-language control over emotion, pacing, whispers, laughter, and pauses — even within a single sentence. Create multi-character scenes, audiobooks, game dialogue, and narration across 83 languages. Available on the web, mobile, and API, with open weights for self-hosting under Fish Audio's community license.

Your recordings should stay with you. Your AI should be your choice. Record on your iPhone, Apple Watch, or recorder. SaidTrace brings your audio to your Mac for transcripts and notes. Choose your AI model, ask questions, and trace answers back to the original recording.

Put GPT, Gemini, Grok and your own agents in one live voice room with your team. They answer, push back on each other and never talk over you. A Scribe writes the decisions. Make the room public and anyone can listen, chat or take the stage. Works in your browser and Google Meet. Free to start: 2,000 credits, no card.