GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
nileagi
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
nileagi
Model Tree
Input Modalities
Output Modalities
Supported Functionality
One short clip (24 kHz). The reference line is the text that was spoken.
| File | Reference (Swahili) |
|---|---|
samples/sample_01.wav | Habari za asubuhi. Karibu kwenye sauti hii. |
python
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessorimport soundfile as sf repo = "nileagi/nileagi-sw-stt"processor = AutoProcessor.from_pretrained(repo)model = AutoModelForSpeechSeq2Seq.from_pretrained(repo)audio, sr = sf.read("samples/sample_01.wav")Prefer 16 kHz mono. Files longer than 25 seconds should be split; the NileAGI service on port 8094 does that.
| Task | Automatic speech recognition |
| Language | Swahili (sw) |
| Direction | Swahili speech → Swahili text |
| Selection | Internal bake-off on held-out local clips |
| Continue-trained on local audio | No |
| Primary metrics | WER, CER |
Held-out clips whose text matches the audio. The companion voice prompt is excluded. Scripture slices are excluded because their stored text does not match the audio.
| Source | n | WER | CER |
|---|---|---|---|
| Conversation | 24 | 13.1% | 2.9% |
| Narration | 8 | 26.5% | 5.3% |
A 64 second paragraph from the companion voice is read back at WER 9.5% and CER 5.4%.
HTTP API (port 8094):
bash
curl -s -F file=@clip.wav http://127.0.0.1:8094/v1/transcribeLibrary:
python
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor processor = AutoProcessor.from_pretrained("nileagi/nileagi-sw-stt")model = AutoModelForSpeechSeq2Seq.from_pretrained("nileagi/nileagi-sw-stt")nileagi/nileagi-sw-ttsnileagi/nileagi-suk-sttnileagi/nileagi-suk-mtTECHNICAL_REPORT.texWeights are redistributed under the MIT License that accompanied the selected public checkpoint. NileAGI evaluated and packaged the checkpoint. That MIT License, not a NileAGI share-alike licence, governs the weights.
See LICENSE and NOTICE.md.
bibtex
@misc{nileagi-sw-stt-2026, title = {Swahili speech-to-text}, author = {Zephania and Isack Odero}, year = {2026}, howpublished = {Hugging Face}, url = {https://huggingface.co/nileagi/nileagi-sw-stt}, note = {NileAGI}}from transformers import AutoModelForSpeechSeq2Seq, AutoProcessorimport soundfile as sf repo = "nileagi/nileagi-sw-stt"processor = AutoProcessor.from_pretrained(repo)model = AutoModelForSpeechSeq2Seq.from_pretrained(repo)audio, sr = sf.read("samples/sample_01.wav")from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor processor = AutoProcessor.from_pretrained("nileagi/nileagi-sw-stt")model = AutoModelForSpeechSeq2Seq.from_pretrained("nileagi/nileagi-sw-stt")@misc{nileagi-sw-stt-2026, title = {Swahili speech-to-text}, author = {Zephania and Isack Odero}, year = {2026}, howpublished = {Hugging Face}, url = {https://huggingface.co/nileagi/nileagi-sw-stt}, note = {NileAGI}}