Alternatives to Kaldi

Gladia

Gladia suits developers building audio recording, transcription, or analysis into products through an API.

Otter.ai

Otter.ai suits people and teams who want searchable meeting transcripts, summaries and action items.

Speechmatics

Speechmatics suits teams building transcription, voice-agent, or speech-synthesis workflows, including listed healthcare, legal, contact center, and meeting use cases.

AssemblyAI

It suits developers building transcription, speech-understanding, or voice applications who need API access, language coverage, or self-hosting.

ESPnet

ESPnet suits developers and researchers building speech-processing workflows who can work with its installation and task dependencies.

Superwhisper

Superwhisper suits people who dictate across apps and want text adapted to screen context.

Whisper

Whisper suits developers and researchers building speech-processing applications or conducting further research.

Speech-to-Text suits developers integrating transcription into applications or workflows that need streaming, batch, or synchronous recognition.

Leopard suits developers and enterprises that need local transcription of recorded audio or video, including word-level metadata and speaker tags.

CMU Sphinx suits developers applying speech technology in applications, rather than speech-recognition researchers.

Voice In suits people who want to dictate into browser-based text fields, use spoken editing commands, and work in supported browsers.

Talon suits people seeking hands-free input for writing, coding, browsing, or gaming.

Vosk suits developers building applications that need offline or streaming speech recognition.

Trint suits newsrooms, podcasters, local businesses, and global organizations that need searchable transcripts, editing, collaboration, or publishing tools.

It suits Windows users and organizations needing dictation, audio transcription, custom commands, or centrally distributed language materials.

It suits developers and teams needing speech synthesis through a web service or API, including voice cloning and several audio formats.