Model card: concise training description

627d25c verified 29 days ago

7.03 kB

---
language:
- am
library_name: nemo
pipeline_tag: automatic-speech-recognition
license: other
license_name: openmdw-1.1
license_link: https://openmdw.ai/license/1-1/
base_model: nvidia/nemotron-3.5-asr-streaming-0.6b
tags:
- speech
- asr
- amharic
- streaming
- realtime
- addis-ai
---
# Addis Scribe Streaming

**Realtime Amharic speech recognition by Addis AI.** Text appears while people speak, in 0.32-second steps,
and is final about 30 ms after they stop. Built on NVIDIA Nemotron 3.5 ASR Streaming (0.6B), trained on
5,000+ hours of quality-checked Amharic speech.

| | Addis Scribe Streaming | [Google Chirp 3](https://cloud.google.com/speech-to-text/v2/docs/chirp_3-model) (streaming) |
|---|---|---|
| Word error rate, natural speech | **28.4%** | 32.5% |
| Character error rate, natural speech | **13.1%** | 16.5% |
| Final text after the speaker stops | **30 ms** | 1,411 ms |
| First words on screen | **1.56 s** | 6.67 s |

## Streaming support

"Streaming" here means the model reads each new piece of audio once, keeps what it learned in a cache, and
updates the text without re-reading earlier audio. Cost and delay stay flat however long someone talks.

| System | Streaming | How it handles live audio |
|---|---|---|
| **Addis Scribe Streaming** | **Native** | Cache-aware: each 0.32 s (or 1.12 s) chunk is processed once. |
| [Google Chirp 3](https://cloud.google.com/speech-to-text/v2/docs/chirp_3-model) | Native (cloud API) | Server-side streaming over the network. |
| [Shook](https://huggingface.co/b1n1yam/shook-medium-amharic-2k) (Whisper medium) | Not supported | Whisper models transcribe complete audio, not a live stream. Our benchmark runs it live using the method of [WhisperLiveKit](https://github.com/QuentinFuxa/WhisperLiveKit), a widely used open-source tool for live Whisper: once per second it transcribes all audio received so far and displays a word once two consecutive transcriptions contain it. Because each pass covers the whole recording so far, the delay grows the longer someone speaks: measured median 10.8 s after speech ends. |
| [Hohe](https://huggingface.co/snapwre/hohe-asr-amharic) (wav2vec2-BERT) | No | Offline only: needs the whole recording before it can answer. |

## Benchmarks

Every system is scored with the same text normaliser (Unicode NFC, punctuation and symbols removed).
WER = word error rate, CER = character error rate; lower is better.

### Natural speech: 100 [WAXAL](https://huggingface.co/datasets/google/WaxalNLP) Amharic test clips

22 speakers, 29 minutes, human transcripts. None of it was used in training. Clips were picked at random
(fixed seed) from WAXAL `amh` test, 2 to 30 s long.

| System | Mode | WER | CER | Final text after speech ends | First words |
|---|---|---|---|---|---|
| **Addis Scribe Streaming** | **streaming 0.32 s** | **28.4%** | **13.1%** | **30 ms** | **1.56 s** |
| Addis Scribe Streaming | streaming 1.12 s | 27.3% | 12.4% | 34 ms | 2.21 s |
| [Google Chirp 3](https://cloud.google.com/speech-to-text/v2/docs/chirp_3-model) (am-ET) | streaming, realtime | 32.5% | 16.5% | 1,411 ms | 6.67 s |
| Google Chirp 3 (am-ET) | batch, whole file | 29.2% | 13.5% | n/a | n/a |
| [Shook](https://huggingface.co/b1n1yam/shook-medium-amharic-2k) (Whisper medium) | live, WhisperLiveKit method (1 s updates) | 35.0% | 16.6% | 10,839 ms | 3.66 s |
| Shook (Whisper medium) | offline | 35.0% | 16.6% | n/a | n/a |
| [Hohe](https://huggingface.co/snapwre/hohe-asr-amharic) (wav2vec2-BERT) | offline only | 25.4% | 11.2% | n/a | n/a |

Latency for Addis Scribe and Shook is measured on one NVIDIA A100 with no network hop, with audio fed at real-time speed; Google's figures
include the network round trip to its EU endpoint from the benchmark machine.

### [FLEURS](https://huggingface.co/datasets/google/fleurs) Amharic test (read sentences)

| System | Mode | WER | CER |
|---|---|---|---|
| **Addis Scribe Streaming** | **streaming 0.32 s** (first 150 clips) | **19.8%** | 8.2% |
| Hohe | offline (same 150 clips) | 20.6% | 7.9% |
| Addis Scribe Streaming | offline (all 516 clips) | 20.6% | 7.7% |
| Hohe | offline (all 516 clips) | 20.2% | 7.3% |
| Previous Addis AI Nemotron | streaming 0.32 s (first 150 clips) | 35.0% | 17.0% |

Every per-clip transcript, latency and the scoring script are in [`benchmark/`](benchmark/), so these
numbers can be reproduced exactly.

## Quickstart

```python
from stream_infer import AmharicStreamer
scribe = AmharicStreamer("addis-scribe-streaming.nemo", chunk="320ms")   # or "1120ms"
for frame in microphone_frames():        # float32 mono, 16 kHz, any frame size
    print(scribe.feed(frame))            # transcript so far
final_text = scribe.finish()             # at end of speech; scribe.reset() for the next utterance
```

Command line: `python stream_infer.py addis-scribe-streaming.nemo speech_16k.wav [320ms|1120ms]`.
Requires [NVIDIA NeMo](https://github.com/NVIDIA/NeMo) 3.0 with `numba-cuda[cu12]`, `numpy<2.4` and `nvidia-nvjitlink-cu12>=12.9`. Fed 100 ms frames
like a live microphone, `AmharicStreamer` reproduces the reference streaming transcripts exactly (100 of 100
benchmark clips).

## Model

| | |
|---|---|
| Architecture | Cache-aware FastConformer encoder, hybrid RNNT + CTC, 0.6B parameters |
| Decoding | Greedy RNNT; language prompt `am-ET` (index 49) |
| Chunk sizes | 0.32 s (default) or 1.12 s |
| Input / output | 16 kHz mono audio / Amharic text without punctuation |
| Files | `addis-scribe-streaming.nemo`, `serving.json`, `stream_infer.py`, `SHA256SUMS` |

## Training

Trained on 5,000+ hours of quality-checked Amharic speech, then refined on human-transcribed speech.
FLEURS, WAXAL test and the evaluation speakers were never used in training.

## Limits

- On natural conversational speech it is about 3 WER points behind Hohe, which is offline only.
- No punctuation (training text had punctuation removed).
- Tested on Amharic only; other languages are out of scope.

## References

| | |
|---|---|
| Base model | [NVIDIA Nemotron 3.5 ASR Streaming 0.6B](https://huggingface.co/nvidia/nemotron-3.5-asr-streaming-0.6b) |
| Toolkit | [NVIDIA NeMo](https://github.com/NVIDIA/NeMo) |
| Google Chirp 3 | [Chirp 3 model documentation](https://cloud.google.com/speech-to-text/v2/docs/chirp_3-model) |
| Hohe | [snapwre/hohe-asr-amharic](https://huggingface.co/snapwre/hohe-asr-amharic) |
| Shook | [b1n1yam/shook-medium-amharic-2k](https://huggingface.co/b1n1yam/shook-medium-amharic-2k) |
| WhisperLiveKit | [QuentinFuxa/WhisperLiveKit](https://github.com/QuentinFuxa/WhisperLiveKit) |
| WAXAL (natural-speech test set) | [google/WaxalNLP](https://huggingface.co/datasets/google/WaxalNLP), `amh` test split, CC-BY-SA-4.0 |
| FLEURS (read-speech test set) | [google/fleurs](https://huggingface.co/datasets/google/fleurs), `am_et` test split |

## License

Built on [NVIDIA Nemotron 3.5 ASR Streaming 0.6B](https://huggingface.co/nvidia/nemotron-3.5-asr-streaming-0.6b),
used under the [OpenMDW-1.1](https://openmdw.ai/license/1-1/) license. © Addis AI.

	---
	language:
	- am
	library_name: nemo
	pipeline_tag: automatic-speech-recognition
	license: other
	license_name: openmdw-1.1
	license_link: https://openmdw.ai/license/1-1/
	base_model: nvidia/nemotron-3.5-asr-streaming-0.6b
	tags:
	- speech
	- asr
	- amharic
	- streaming
	- realtime
	- addis-ai
	---

	# Addis Scribe Streaming

	Realtime Amharic speech recognition by Addis AI. Text appears while people speak, in 0.32-second steps,
	and is final about 30 ms after they stop. Built on NVIDIA Nemotron 3.5 ASR Streaming (0.6B), trained on
	5,000+ hours of quality-checked Amharic speech.

	\| \| Addis Scribe Streaming \| [Google Chirp 3](https://cloud.google.com/speech-to-text/v2/docs/chirp_3-model) (streaming) \|
	\|---\|---\|---\|
	\| Word error rate, natural speech \| 28.4% \| 32.5% \|
	\| Character error rate, natural speech \| 13.1% \| 16.5% \|
	\| Final text after the speaker stops \| 30 ms \| 1,411 ms \|
	\| First words on screen \| 1.56 s \| 6.67 s \|

	## Streaming support

	"Streaming" here means the model reads each new piece of audio once, keeps what it learned in a cache, and
	updates the text without re-reading earlier audio. Cost and delay stay flat however long someone talks.

	\| System \| Streaming \| How it handles live audio \|
	\|---\|---\|---\|
	\| Addis Scribe Streaming \| Native \| Cache-aware: each 0.32 s (or 1.12 s) chunk is processed once. \|
	\| [Google Chirp 3](https://cloud.google.com/speech-to-text/v2/docs/chirp_3-model) \| Native (cloud API) \| Server-side streaming over the network. \|
	\| [Shook](https://huggingface.co/b1n1yam/shook-medium-amharic-2k) (Whisper medium) \| Not supported \| Whisper models transcribe complete audio, not a live stream. Our benchmark runs it live using the method of [WhisperLiveKit](https://github.com/QuentinFuxa/WhisperLiveKit), a widely used open-source tool for live Whisper: once per second it transcribes all audio received so far and displays a word once two consecutive transcriptions contain it. Because each pass covers the whole recording so far, the delay grows the longer someone speaks: measured median 10.8 s after speech ends. \|
	\| [Hohe](https://huggingface.co/snapwre/hohe-asr-amharic) (wav2vec2-BERT) \| No \| Offline only: needs the whole recording before it can answer. \|

	## Benchmarks

	Every system is scored with the same text normaliser (Unicode NFC, punctuation and symbols removed).
	WER = word error rate, CER = character error rate; lower is better.

	### Natural speech: 100 [WAXAL](https://huggingface.co/datasets/google/WaxalNLP) Amharic test clips

	22 speakers, 29 minutes, human transcripts. None of it was used in training. Clips were picked at random
	(fixed seed) from WAXAL `amh` test, 2 to 30 s long.

	\| System \| Mode \| WER \| CER \| Final text after speech ends \| First words \|
	\|---\|---\|---\|---\|---\|---\|
	\| Addis Scribe Streaming \| streaming 0.32 s \| 28.4% \| 13.1% \| 30 ms \| 1.56 s \|
	\| Addis Scribe Streaming \| streaming 1.12 s \| 27.3% \| 12.4% \| 34 ms \| 2.21 s \|
	\| [Google Chirp 3](https://cloud.google.com/speech-to-text/v2/docs/chirp_3-model) (am-ET) \| streaming, realtime \| 32.5% \| 16.5% \| 1,411 ms \| 6.67 s \|
	\| Google Chirp 3 (am-ET) \| batch, whole file \| 29.2% \| 13.5% \| n/a \| n/a \|
	\| [Shook](https://huggingface.co/b1n1yam/shook-medium-amharic-2k) (Whisper medium) \| live, WhisperLiveKit method (1 s updates) \| 35.0% \| 16.6% \| 10,839 ms \| 3.66 s \|
	\| Shook (Whisper medium) \| offline \| 35.0% \| 16.6% \| n/a \| n/a \|
	\| [Hohe](https://huggingface.co/snapwre/hohe-asr-amharic) (wav2vec2-BERT) \| offline only \| 25.4% \| 11.2% \| n/a \| n/a \|

	Latency for Addis Scribe and Shook is measured on one NVIDIA A100 with no network hop, with audio fed at real-time speed; Google's figures
	include the network round trip to its EU endpoint from the benchmark machine.

	### [FLEURS](https://huggingface.co/datasets/google/fleurs) Amharic test (read sentences)

	\| System \| Mode \| WER \| CER \|
	\|---\|---\|---\|---\|
	\| Addis Scribe Streaming \| streaming 0.32 s (first 150 clips) \| 19.8% \| 8.2% \|
	\| Hohe \| offline (same 150 clips) \| 20.6% \| 7.9% \|
	\| Addis Scribe Streaming \| offline (all 516 clips) \| 20.6% \| 7.7% \|
	\| Hohe \| offline (all 516 clips) \| 20.2% \| 7.3% \|
	\| Previous Addis AI Nemotron \| streaming 0.32 s (first 150 clips) \| 35.0% \| 17.0% \|

	Every per-clip transcript, latency and the scoring script are in [`benchmark/`](benchmark/), so these
	numbers can be reproduced exactly.

	## Quickstart

	```python
	from stream_infer import AmharicStreamer

	scribe = AmharicStreamer("addis-scribe-streaming.nemo", chunk="320ms") # or "1120ms"
	for frame in microphone_frames(): # float32 mono, 16 kHz, any frame size
	print(scribe.feed(frame)) # transcript so far
	final_text = scribe.finish() # at end of speech; scribe.reset() for the next utterance
	```

	Command line: `python stream_infer.py addis-scribe-streaming.nemo speech_16k.wav [320ms\|1120ms]`.
	Requires [NVIDIA NeMo](https://github.com/NVIDIA/NeMo) 3.0 with `numba-cuda[cu12]`, `numpy<2.4` and `nvidia-nvjitlink-cu12>=12.9`. Fed 100 ms frames
	like a live microphone, `AmharicStreamer` reproduces the reference streaming transcripts exactly (100 of 100
	benchmark clips).

	## Model

	\| \| \|
	\|---\|---\|
	\| Architecture \| Cache-aware FastConformer encoder, hybrid RNNT + CTC, 0.6B parameters \|
	\| Decoding \| Greedy RNNT; language prompt `am-ET` (index 49) \|
	\| Chunk sizes \| 0.32 s (default) or 1.12 s \|
	\| Input / output \| 16 kHz mono audio / Amharic text without punctuation \|
	\| Files \| `addis-scribe-streaming.nemo`, `serving.json`, `stream_infer.py`, `SHA256SUMS` \|

	## Training

	Trained on 5,000+ hours of quality-checked Amharic speech, then refined on human-transcribed speech.
	FLEURS, WAXAL test and the evaluation speakers were never used in training.

	## Limits

	- On natural conversational speech it is about 3 WER points behind Hohe, which is offline only.
	- No punctuation (training text had punctuation removed).
	- Tested on Amharic only; other languages are out of scope.

	## References

	\| \| \|
	\|---\|---\|
	\| Base model \| [NVIDIA Nemotron 3.5 ASR Streaming 0.6B](https://huggingface.co/nvidia/nemotron-3.5-asr-streaming-0.6b) \|
	\| Toolkit \| [NVIDIA NeMo](https://github.com/NVIDIA/NeMo) \|
	\| Google Chirp 3 \| [Chirp 3 model documentation](https://cloud.google.com/speech-to-text/v2/docs/chirp_3-model) \|
	\| Hohe \| [snapwre/hohe-asr-amharic](https://huggingface.co/snapwre/hohe-asr-amharic) \|
	\| Shook \| [b1n1yam/shook-medium-amharic-2k](https://huggingface.co/b1n1yam/shook-medium-amharic-2k) \|
	\| WhisperLiveKit \| [QuentinFuxa/WhisperLiveKit](https://github.com/QuentinFuxa/WhisperLiveKit) \|
	\| WAXAL (natural-speech test set) \| [google/WaxalNLP](https://huggingface.co/datasets/google/WaxalNLP), `amh` test split, CC-BY-SA-4.0 \|
	\| FLEURS (read-speech test set) \| [google/fleurs](https://huggingface.co/datasets/google/fleurs), `am_et` test split \|

	## License

	Built on [NVIDIA Nemotron 3.5 ASR Streaming 0.6B](https://huggingface.co/nvidia/nemotron-3.5-asr-streaming-0.6b),
	used under the [OpenMDW-1.1](https://openmdw.ai/license/1-1/) license. © Addis AI.