IMS-Speech: A Speech to Text Tool
About
We present the IMS-Speech, a web based tool for German and English speech transcription aiming to facilitate research in various disciplines which require accesses to lexical information in spoken language materials. This tool is based on modern open source software stack, advanced speech recognition methods and public data resources and is freely available for academic researchers. The utilized models are built to be generic in order to provide transcriptions of competitive accuracy on a diverse set of tasks and conditions.
Pavel Denisov, Ngoc Thang Vu• 2019
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Automatic Speech Recognition | LibriSpeech (test-other) | WER12.5 | 966 | |
| Automatic Speech Recognition | LibriSpeech clean (test) | WER4.3 | 833 | |
| Speech Recognition | WSJ (92-eval) | WER3.8 | 131 | |
| Automatic Speech Recognition | Tuda-DE (test) | WER10 | 26 | |
| Automatic Speech Recognition | Tuda-DE (dev) | WER11.1 | 24 | |
| Automatic Speech Recognition | AMI SDM English (eval) | WER38.5 | 8 | |
| Automatic Speech Recognition | Verbmobil 1 German (test) | WER7.3 | 4 | |
| Automatic Speech Recognition | AMI IHM English (eval) | WER17.4 | 2 | |
| Automatic Speech Recognition | AMI MDM English (eval) | WER34.1 | 2 | |
| Automatic Speech Recognition | Verbmobil German 1 (dev) | WER0.067 | 2 |
Showing 10 of 12 rows