Speech

Speechlab localizes audio and video with dubbing, captions, and subtitles in more than 50 languages.

OpenUtau is an open-source editor for the UTAU community and singing synthesis, with setup guides and contribution instructions.

VoicePal reads documents, text and web articles aloud using offline voices in multiple languages, with apps listed for Windows, Mac and Android.

EmotiVoice provides a multi-voice text-to-speech engine with prompt controls.

ElevenReader reads text, ebooks, PDFs, and audiobooks aloud with text to speech voices.

Text-to-Audio Generation with Latent Diffusion Models - Speech Research.

VanillaVoice reads text aloud using a selection of human sounding voices.

Sonix transcribes audio and video in more than 54 languages, with AI summaries and editing tools.

VEED generates subtitles for video and audio, with language and social video options.

PlainScribe transcribes and translates audio or video into 47 languages and can produce summaries and subtitles.

Translate Shell offers command-line translation through Google, Bing, and Yandex services.

VoicePen turns meetings, lectures, memos, videos and podcasts into transcripts, notes and summaries, with apps for iPhone, Watch, iPad and Mac.

Read Aloud reads webpage content aloud and lists browser extensions for Chrome, Firefox, and Edge.

Fish Audio offers AI text-to-speech and voice cloning with controls for generated emotion.

Smallest AI provides APIs for text-to-speech, speech-to-text, and voice agents.

Sayboard uses Vosk for on-device speech recognition in an open source Android voice keyboard.

WellSaid generates voiceovers with AI voices in different dialects and production styles.

NeMo Parakeet ASR Models attain strong speech recognition accuracy while being efficient for inference. Available in CTC and RNN-Transducer variants.

KubeAI serves machine learning models on Kubernetes, including speech to text, language, vision, and embedding models.

HushMic suppresses microphone noise in real time on Linux using a system-wide virtual microphone with PipeWire.

AnyVoiceLab generates speech from a short voice sample and lets users download the resulting audio without logging in.

Voxify generates speech from text with more than 450 voice options and controls for pitch, speed and emotion.

VoxWild runs offline on Windows for text-to-speech, with 13 voices, cloning from short recordings and batch generation.

AnyVoiceLab converts long passages of text into speech, with support for up to 50,000 characters.