# Speaker Diarization on CALLHOME American English 2-speaker (CH-109) (eval)

Source: [Speaker Diarization on CALLHOME American English 2-speaker (CH-109) (eval)](<https://www.sota2.com/research/sota/speaker-diarization-on-callhome-american-english-2-speaker-ch-109-eval>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:39:45.771Z

Evaluation of Speaker Diarization performance on the evaluation split of the CALLHOME American English 2-speaker (CH-109) dataset.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Speaker Diarization |
| Dataset (including any stated split) | CALLHOME American English 2-speaker (CH-109) (eval) |
| Metric | Confusion |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 2 of 2 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Confusion](<https://www.sota2.com/research/sota/speaker-diarization-on-callhome-american-english-2-speaker-ch-109-eval.md?metric=Confusion>): Lower is better; unit: PERCENT.

- [FA](<https://www.sota2.com/research/sota/speaker-diarization-on-callhome-american-english-2-speaker-ch-109-eval.md?metric=FA>): Lower is better; unit: PERCENT.

- [Miss Rate](<https://www.sota2.com/research/sota/speaker-diarization-on-callhome-american-english-2-speaker-ch-109-eval.md?metric=Miss%20Rate>): Lower is better; unit: PERCENT.

- [Total DER](<https://www.sota2.com/research/sota/speaker-diarization-on-callhome-american-english-2-speaker-ch-109-eval.md?metric=Total%20DER>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Confusion | Result date | Report paper |
| --- | --- | --- | --- |
| d-vector + spectral | 5.97 | Not recorded | 68fa6ba61e209131cea2f934 |
| Zajíc et al. | 7.84 | Not recorded | 68fa6ba61e209131cea2f934 |

### d-vector + spectral

Confusion: 5.97. Result ID: 698e9c1affc3901de09f1cd6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ba61e209131cea2f934.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ba61e209131cea2f934.md>)

[Implementation](<https://google.github.io/speaker-id/publications/LstmDiarization>)

Recorded evaluation configuration: {"embedding":"d-vector","clustering":"spectral"}. This record does not assert independent reproduction.

### Zajíc et al.

Confusion: 7.84. Result ID: 698e9c1affc3901de09f1cd7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ba61e209131cea2f934.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.