Frequently asked questions

Most ASR providers are running at 90-95% margins. Fennec slashes this down to provide you the lowest prices possible. Fennec is based on a fine-tuned open source ASR model and blended with a lightweight LLM recontextualizer that offers some of the lowest WER on the market, while being extremely affordable.

Our namesake is the light and nimble Fennec Fox, which is known for its incredible speed and ultra sensitive hearing.

Fennec supports the following audio formats:

.wav, .aif, .aiff, .au, .snd, .flac, .m4a, .caf, .aac, .mp4, .mp3, .ogg, .oga, .opus, .wma, .ac3, .eac3, .dts, .amr, .gsm, .spx, .mpc, .ape, .aa3, .oma, .sph, .ulaw, .alaw, .g722, .g726, .dsf, .dff, .raw, .pcm, .s16, .m4v, .mov, .mkv, .mka, .webm, .avi, .mpg, .mpeg, .ts, .m2ts, .flv, .3gp, .3g2, .ogv, .asf, .wmv, .mxf

Your data is immediately deleted from our servers upon job completion, but the transcripts persist so you can retrieve them any time.

Yes. To keep things lightweight (we're a small team) we haven't jumped on creating a Node SDK. However, please check out our Python SDK here.

Yes! If you're a startup, please contact us and you'll get 500 hours on us.

We have a LOT in the pipeline right now. Diarization, podcast mode, and WebRTC compatibility are the priorities right now.